Compare commits

...

45 Commits

Author SHA1 Message Date
João Moura
82f9b26848 Merge branch 'main' into devin/1735488202-add-tool-documentation 2024-12-31 01:52:01 -03:00
Marco Vinciguerra
a548463fae feat: add docstring (#1819)
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-31 01:51:43 -03:00
devin-ai-integration[bot]
45b802a625 Docstring, Error Handling, and Type Hints Improvements (#1828)
* docs: add comprehensive docstrings to Flow class and methods

- Added NumPy-style docstrings to all decorator functions
- Added detailed documentation to Flow class methods
- Included parameter types, return types, and examples
- Enhanced documentation clarity and completeness

Co-Authored-By: Joe Moura <joao@crewai.com>

* feat: add secure path handling utilities

- Add path_utils.py with safe path handling functions
- Implement path validation and security checks
- Integrate secure path handling in flow_visualizer.py
- Add path validation in html_template_handler.py
- Add comprehensive error handling for path operations

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: add comprehensive docstrings and type hints to flow utils (#1819)

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: add type annotations and fix import sorting

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: add type annotations to flow utils and visualization utils

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: resolve import sorting and type annotation issues

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: properly initialize and update edge_smooth variable

Co-Authored-By: Joe Moura <joao@crewai.com>

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-31 01:39:19 -03:00
devin-ai-integration[bot]
ba0965ef87 fix: add tiktoken as explicit dependency and document Rust requirement (#1826)
* feat: add tiktoken as explicit dependency and document Rust requirement

- Add tiktoken>=0.8.0 as explicit dependency to ensure pre-built wheels are used
- Document Rust compiler requirement as fallback in README.md
- Addresses issue #1824 tiktoken build failure

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: adjust tiktoken version to ~=0.7.0 for dependency compatibility

- Update tiktoken dependency to ~=0.7.0 to resolve conflict with embedchain
- Maintain compatibility with crewai-tools dependency chain
- Addresses CI build failures

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: add troubleshooting section and make tiktoken optional

Co-Authored-By: Joe Moura <joao@crewai.com>

* Update README.md

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-30 17:10:56 -03:00
devin-ai-integration[bot]
d85898cf29 fix(manager_llm): handle coworker role name case/whitespace properly (#1820)
* fix(manager_llm): handle coworker role name case/whitespace properly

- Add .strip() to agent name and role comparisons in base_agent_tools.py
- Add test case for varied role name cases and whitespace
- Fix issue #1503 with manager LLM delegation

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix(manager_llm): improve error handling and add debug logging

- Add debug logging for better observability
- Add sanitize_agent_name helper method
- Enhance error messages with more context
- Add parameterized tests for edge cases:
  - Embedded quotes
  - Trailing newlines
  - Multiple whitespace
  - Case variations
  - None values
- Improve error handling with specific exceptions

Co-Authored-By: Joe Moura <joao@crewai.com>

* style: fix import sorting in base_agent_tools and test_manager_llm_delegation

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix(manager_llm): improve whitespace normalization in role name matching

Co-Authored-By: Joe Moura <joao@crewai.com>

* style: fix import sorting in base_agent_tools and test_manager_llm_delegation

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix(manager_llm): add error message template for agent tool execution errors

Co-Authored-By: Joe Moura <joao@crewai.com>

* style: fix import sorting in test_manager_llm_delegation.py

Co-Authored-By: Joe Moura <joao@crewai.com>

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-30 16:58:18 -03:00
Devin AI
09fd6058b0 Add comprehensive documentation for all tools
- Added documentation for file operation tools
- Added documentation for search tools
- Added documentation for web scraping tools
- Added documentation for specialized tools (RAG, code interpreter)
- Added documentation for API-based tools (SerpApi, Serply)

Link to Devin run: https://app.devin.ai/sessions/d2f72a2dfb214659aeb3e9f67ed961f7

Co-Authored-By: Joe Moura <joao@crewai.com>
2024-12-29 16:03:22 +00:00
devin-ai-integration[bot]
73f328860b Fix interpolation for output_file in Task (#1803) (#1814)
* fix: interpolate output_file attribute from YAML

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: add security validation for output_file paths

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: add _original_output_file private attribute to fix type-checker error

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: update interpolate_only to handle None inputs and remove duplicate attribute

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: improve output_file validation and error messages

Co-Authored-By: Joe Moura <joao@crewai.com>

* test: add end-to-end tests for output_file functionality

Co-Authored-By: Joe Moura <joao@crewai.com>

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-29 01:57:59 -03:00
João Moura
a0c322a535 fixing file paths for knowledge source 2024-12-28 02:05:19 -03:00
devin-ai-integration[bot]
86f58c95de docs: add agent-specific knowledge documentation and examples (#1811)
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-28 01:48:51 -03:00
João Moura
99fe91586d Update README.md 2024-12-28 01:03:33 -03:00
devin-ai-integration[bot]
0c2d23dfe0 docs: update README to highlight Flows (#1809)
* docs: highlight Flows feature in README

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: enhance README with LangGraph comparison and flows-crews synergy

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: replace initial Flow example with advanced Flow+Crew example; enhance LangGraph comparison

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: incorporate key terms and enhance feature descriptions

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: refine technical language, enhance feature descriptions, fix string interpolation

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: update README with performance metrics, feature enhancements, and course links

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: update LangGraph comparison with paragraph and P.S. section

Co-Authored-By: Joe Moura <joao@crewai.com>

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-28 01:00:58 -03:00
devin-ai-integration[bot]
2433819c4f fix: handle optional storage with null checks (#1808)
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-27 21:30:39 -03:00
Erick Amorim
97fc44c930 fix: Change storage initialization to None for KnowledgeStorage (#1804)
* fix: Change storage initialization to None for KnowledgeStorage

* refactor: Change storage field to optional and improve error handling when saving documents

---------

Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-27 21:18:25 -03:00
siddharth Sambharia
409892d65f Portkey Integration with CrewAI (#1233)
* Create Portkey-Observability-and-Guardrails.md

* crewAI update with new changes

* small change

---------

Co-authored-by: siddharthsambharia-portkey <siddhath.s@portkey.ai>
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-27 18:16:47 -03:00
devin-ai-integration[bot]
62f3df7ed5 docs: add guide for multimodal agents (#1807)
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-27 18:16:02 -03:00
João Igor
4cf8913d31 chore: removing crewai-tools from dev-dependencies (#1760)
As mentioned in issue #1759, listing crewai-tools as dev-dependencies makes pip install it a required dependency, and not an optional

Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-27 17:45:06 -03:00
João Moura
82647358b2 Adding Multimodal Abilities to Crew (#1805)
* initial fix on delegation tools

* fixing tests for delegations and coding

* Refactor prepare tool and adding initial add images logic

* supporting image tool

* fixing linter

* fix linter

* Making sure multimodal feature support i18n

* fix linter and types

* mixxing translations

* fix types and linter

* Revert "fixing linter"

This reverts commit 2eda5fdeed.

* fix linters

* test

* fix

* fix

* fix linter

* fix

* ignore

* type improvements
2024-12-27 17:03:35 -03:00
Brandon Hancock (bhancock_ai)
6cc2f510bf Feat/joao flow improvement requests (#1795)
* Add in or and and in router

* In the middle of improving plotting

* final plot changes

---------

Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-24 18:55:44 -03:00
Lorenze Jay
9a65abf6b8 removed some redundancies (#1796)
* removed some redundancies

* cleanup
2024-12-23 13:54:16 -05:00
Lorenze Jay
b3185ad90c Feat/docling-support (#1763)
* added tool for docling support

* docling support installation

* use file_paths instead of file_path

* fix import

* organized imports

* run_type docs

* needs to be list

* fixed logic

* logged but file_path is backwards compatible

* use file_paths instead of file_path 2

* added test for multiple sources for file_paths

* fix run-types

* enabling local files to work and type cleanup

* linted

* fix test and types

* fixed run types

* fix types

* renamed to CrewDoclingSource

* linted

* added docs

* resolve conflicts

---------

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
Co-authored-by: Brandon Hancock <brandon@brandonhancock.io>
2024-12-23 13:19:58 -05:00
devin-ai-integration[bot]
c887ff1f47 feat: Add interpolate_only method and improve error handling (#1791)
* Fixed output_file not respecting system path

* Fixed yaml config is not escaped properly for output requirements

* feat: Add interpolate_only method and improve error handling

- Add interpolate_only method for string interpolation while preserving JSON structure
- Add comprehensive test coverage for interpolate_only
- Add proper type annotation for logger using ClassVar
- Improve error handling and documentation for _save_file method

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Sort imports to fix lint issues

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Reorganize imports using ruff --fix

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Consolidate imports and fix formatting

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Apply ruff automatic import sorting

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Sort imports using ruff --fix

Co-Authored-By: Joe Moura <joao@crewai.com>

---------

Co-authored-by: Frieda (Jingying) Huang <jingyingfhuang@gmail.com>
Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
Co-authored-by: Frieda Huang <124417784+frieda-huang@users.noreply.github.com>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
2024-12-23 13:05:29 -05:00
devin-ai-integration[bot]
22e5d39884 feat: Add task guardrails feature (#1742)
* feat: Add task guardrails feature

Add support for custom code guardrails in tasks that validate outputs
before proceeding to the next task. Features include:

- Optional task-level guardrail function
- Pre-next-task execution timing
- Tuple return format (success, data)
- Automatic result/error routing
- Configurable retry mechanism
- Comprehensive documentation and tests

Link to Devin run: https://app.devin.ai/sessions/39f6cfd6c5a24d25a7bd70ce070ed29a

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Add type check for guardrail result and remove unused import

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Remove unnecessary f-string prefix

Co-Authored-By: Joe Moura <joao@crewai.com>

* feat: Add guardrail validation improvements

- Add result/error exclusivity validation in GuardrailResult
- Make return type annotations optional in Task guardrail validator
- Improve error messages for validation failures

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: Add comprehensive guardrails documentation

- Add type hints and examples
- Add error handling best practices
- Add structured error response patterns
- Document retry mechanisms
- Improve documentation organization

Co-Authored-By: Joe Moura <joao@crewai.com>

* refactor: Update guardrail functions to handle TaskOutput objects

Co-Authored-By: Joe Moura <joao@crewai.com>

* feat: Add task guardrails feature

Add support for custom code guardrails in tasks that validate outputs
before proceeding to the next task. Features include:

- Optional task-level guardrail function
- Pre-next-task execution timing
- Tuple return format (success, data)
- Automatic result/error routing
- Configurable retry mechanism
- Comprehensive documentation and tests

Link to Devin run: https://app.devin.ai/sessions/39f6cfd6c5a24d25a7bd70ce070ed29a

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Add type check for guardrail result and remove unused import

Co-Authored-By: Joe Moura <joao@crewai.com>

* fix: Remove unnecessary f-string prefix

Co-Authored-By: Joe Moura <joao@crewai.com>

* feat: Add guardrail validation improvements

- Add result/error exclusivity validation in GuardrailResult
- Make return type annotations optional in Task guardrail validator
- Improve error messages for validation failures

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: Add comprehensive guardrails documentation

- Add type hints and examples
- Add error handling best practices
- Add structured error response patterns
- Document retry mechanisms
- Improve documentation organization

Co-Authored-By: Joe Moura <joao@crewai.com>

* refactor: Update guardrail functions to handle TaskOutput objects

Co-Authored-By: Joe Moura <joao@crewai.com>

* style: Fix import sorting in task guardrails files

Co-Authored-By: Joe Moura <joao@crewai.com>

* fixing docs

* Fixing guardarils implementation

* docs: Enhance guardrail validator docstring with runtime validation rationale

Co-Authored-By: Joe Moura <joao@crewai.com>

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-12-22 00:52:02 -03:00
PJ
9ee6824ccd Correcting a small grammatical issue that was bugging me: from _satisfy the expect criteria_ to _satisfies the expected criteria_ (#1783)
Signed-off-by: PJ Hagerty <pjhagerty@gmail.com>
Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-20 10:17:34 -05:00
Vini Brasil
da73865f25 Add tool.crewai.type pyproject attribute in templates (#1789) 2024-12-20 10:36:18 -03:00
Vini Brasil
627b9f1abb Remove relative import in flow main.py template (#1782) 2024-12-18 10:47:44 -03:00
alan blount
1b8001bf98 Gemini 2.0 (#1773)
* Update llms.mdx (Gemini 2.0)

- Add Gemini 2.0 flash to Gemini table.
- Add link to 2 hosting paths for Gemini in Tip.
- Change to lower case model slugs vs names, user convenience.
- Add https://artificialanalysis.ai/ as alternate leaderboard.
- Move Gemma to "other" tab.

* Update llm.py (gemini 2.0)

Add setting for Gemini 2.0 context window to llm.py

---------

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-17 16:44:10 -05:00
Tony Kipkemboi
e59e07e4f7 Merge pull request #1777 from crewAIInc/fix/python-max-version
Fix/python max version
2024-12-17 16:09:44 -05:00
Brandon Hancock
ee239b1c06 change to <13 instead of <=12 2024-12-17 16:00:15 -05:00
Brandon Hancock
bf459bf983 include 12 but not 13 2024-12-17 15:29:11 -05:00
Karan Vaidya
94eaa6740e Fix bool and null handling (#1771) 2024-12-16 16:23:53 -05:00
Shahar Yair
6d7c1b0743 Fix: CrewJSONEncoder now accepts enums (#1752)
* bugfix: CrewJSONEncoder now accepts enums

* sort imports

---------

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-12 15:13:10 -05:00
Brandon Hancock (bhancock_ai)
6b864ee21d drop print (#1755) 2024-12-12 15:08:37 -05:00
Brandon Hancock (bhancock_ai)
1ffa8904db apply agent ops changes and resolve merge conflicts (#1748)
* apply agent ops changes and resolve merge conflicts

* Trying to fix tests

* add back in vcr

* update tools

* remove pkg_resources which was causing issues

* Fix tests

* experimenting to see if unique content is an issue with knowledge

* experimenting to see if unique content is an issue with knowledge

* update chromadb which seems to have issues with upsert

* generate new yaml for failing test

* Investigating upsert

* Drop patch

* Update casettes

* Fix duplicate document issue

* more fixes

* add back in vcr

* new cassette for test

---------

Co-authored-by: Lorenze Jay <lorenzejaytech@gmail.com>
2024-12-12 15:04:32 -05:00
Brandon Hancock (bhancock_ai)
ad916abd76 remove pkg_resources which was causing issues (#1751) 2024-12-12 12:41:13 -05:00
Rip&Tear
9702711094 Feature/add workflow permissions (#1749)
* fix: Call ChromaDB reset before removing storage directory to fix disk I/O errors

* feat: add workflow permissions to stale.yml

* revert rag_storage.py changes

* revert rag_storage.py changes

---------

Co-authored-by: Matt B <mattb@Matts-MacBook-Pro.local>
Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-12 12:31:43 -05:00
André Lago
8094754239 Fix small typo in sample tool (#1747)
Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-12 10:11:47 -05:00
Rashmi Pawar
bc5e303d5f NVIDIA Provider : UI changes (#1746)
* docs: add nvidia as provider

* nvidia ui docs changes

* add note for updated list

---------

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-12 10:01:53 -05:00
Anmol Deep
ec89e003c8 Added is_auto_end flag in agentops.end session in crew.py (#1320)
When using agentops, we have the option to pass the `skip_auto_end_session` parameter, which is supposed to not end the session if the `end_session` function is called by Crew.

Now the way it works is, the `agentops.end_session` accepts `is_auto_end` flag and crewai should have passed it as `True` (its `False` by default). 

I have changed the code to pass is_auto_end=True

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-11 11:34:17 -05:00
Bowen Liang
0b0f2d30ab sort imports with isort rules by ruff linter (#1730)
* sort imports

* update

---------

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
Co-authored-by: Eduardo Chiarotti <dudumelgaco@hotmail.com>
2024-12-11 10:46:53 -05:00
Brandon Hancock (bhancock_ai)
1df61aba4c include event emitter in flows (#1740)
* include event emitter in flows

* Clean up

* Fix linter
2024-12-11 10:16:05 -05:00
Paul Cowgill
da9220fa81 Remove manager_callbacks reference (#1741) 2024-12-11 10:13:57 -05:00
Archkon
da4f356fab fix:typo error (#1738)
* Update base_agent_tools.py

typo error

* Update main.py

typo error

* Update base_file_knowledge_source.py

typo error

* Update test_main.py

typo error

* Update en.json

* Update prompts.json

---------

Co-authored-by: Brandon Hancock (bhancock_ai) <109994880+bhancockio@users.noreply.github.com>
2024-12-10 11:18:45 -05:00
Brandon Hancock (bhancock_ai)
d932b20c6e copy googles changes. Fix tests. Improve LLM file (#1737)
* copy googles changes. Fix tests. Improve LLM file

* Fix type issue
2024-12-10 11:14:37 -05:00
Brandon Hancock (bhancock_ai)
2f9a2afd9e Update pyproject.toml and uv.lock to drop crewai-tools as a default requirement (#1711) 2024-12-09 14:17:46 -05:00
Brandon Hancock (bhancock_ai)
c1df7c410e Bugfix/restrict python version compatibility (#1736)
* drop 3.13

* revert

* Drop test cassette that was causing error

* trying to fix failing test

* adding thiago changes

* resolve final tests

* Drop skip

* drop pipeline
2024-12-09 14:07:57 -05:00
145 changed files with 14790 additions and 12217 deletions

View File

@@ -13,4 +13,4 @@ jobs:
pip install ruff
- name: Run Ruff Linter
run: ruff check --exclude "templates","__init__.py"
run: ruff check

View File

@@ -1,5 +1,10 @@
name: Mark stale issues and pull requests
permissions:
contents: write
issues: write
pull-requests: write
on:
schedule:
- cron: '10 12 * * *'
@@ -8,9 +13,6 @@ on:
jobs:
stale:
runs-on: ubuntu-latest
permissions:
issues: write
pull-requests: write
steps:
- uses: actions/stale@v9
with:

View File

@@ -1,9 +1,7 @@
repos:
- repo: https://github.com/astral-sh/ruff-pre-commit
rev: v0.4.4
rev: v0.8.2
hooks:
- id: ruff
args: ["--fix"]
exclude: "templates"
- id: ruff-format
exclude: "templates"

9
.ruff.toml Normal file
View File

@@ -0,0 +1,9 @@
exclude = [
"templates",
"__init__.py",
]
[lint]
select = [
"I", # isort rules
]

177
README.md
View File

@@ -4,7 +4,7 @@
# **CrewAI**
🤖 **CrewAI**: Cutting-edge framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
🤖 **CrewAI**: Production-grade framework for orchestrating sophisticated AI agent systems. From simple automations to complex real-world applications, CrewAI provides precise control and deep customization. By fostering collaborative intelligence through flexible, production-ready architecture, CrewAI empowers agents to work together seamlessly, tackling complex business challenges with predictable, consistent results.
<h3>
@@ -22,13 +22,17 @@
- [Why CrewAI?](#why-crewai)
- [Getting Started](#getting-started)
- [Key Features](#key-features)
- [Understanding Flows and Crews](#understanding-flows-and-crews)
- [CrewAI vs LangGraph](#how-crewai-compares)
- [Examples](#examples)
- [Quick Tutorial](#quick-tutorial)
- [Write Job Descriptions](#write-job-descriptions)
- [Trip Planner](#trip-planner)
- [Stock Analysis](#stock-analysis)
- [Using Crews and Flows Together](#using-crews-and-flows-together)
- [Connecting Your Crew to a Model](#connecting-your-crew-to-a-model)
- [How CrewAI Compares](#how-crewai-compares)
- [Frequently Asked Questions (FAQ)](#frequently-asked-questions-faq)
- [Contribution](#contribution)
- [Telemetry](#telemetry)
- [License](#license)
@@ -36,22 +40,51 @@
## Why CrewAI?
The power of AI collaboration has too much to offer.
CrewAI is designed to enable AI agents to assume roles, share goals, and operate in a cohesive unit - much like a well-oiled crew. Whether you're building a smart assistant platform, an automated customer service ensemble, or a multi-agent research team, CrewAI provides the backbone for sophisticated multi-agent interactions.
CrewAI is a standalone framework, built from the ground up without dependencies on Langchain or other agent frameworks. It's designed to enable AI agents to assume roles, share goals, and operate in a cohesive unit - much like a well-oiled crew. Whether you're building a smart assistant platform, an automated customer service ensemble, or a multi-agent research team, CrewAI provides the backbone for sophisticated multi-agent interactions.
## Getting Started
### Learning Resources
Learn CrewAI through our comprehensive courses:
- [Multi AI Agent Systems with CrewAI](https://www.deeplearning.ai/short-courses/multi-ai-agent-systems-with-crewai/) - Master the fundamentals of multi-agent systems
- [Practical Multi AI Agents and Advanced Use Cases](https://www.deeplearning.ai/short-courses/practical-multi-ai-agents-and-advanced-use-cases-with-crewai/) - Deep dive into advanced implementations
### Understanding Flows and Crews
CrewAI offers two powerful, complementary approaches that work seamlessly together to build sophisticated AI applications:
1. **Crews**: Teams of AI agents with true autonomy and agency, working together to accomplish complex tasks through role-based collaboration. Crews enable:
- Natural, autonomous decision-making between agents
- Dynamic task delegation and collaboration
- Specialized roles with defined goals and expertise
- Flexible problem-solving approaches
2. **Flows**: Production-ready, event-driven workflows that deliver precise control over complex automations. Flows provide:
- Fine-grained control over execution paths for real-world scenarios
- Secure, consistent state management between tasks
- Clean integration of AI agents with production Python code
- Conditional branching for complex business logic
The true power of CrewAI emerges when combining Crews and Flows. This synergy allows you to:
- Build complex, production-grade applications
- Balance autonomy with precise control
- Handle sophisticated real-world scenarios
- Maintain clean, maintainable code structure
### Getting Started with Installation
To get started with CrewAI, follow these simple steps:
### 1. Installation
Ensure you have Python >=3.10 <=3.12 installed on your system. CrewAI uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.
Ensure you have Python >=3.10 <3.13 installed on your system. CrewAI uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.
First, install CrewAI:
```shell
pip install crewai
```
If you want to install the 'crewai' package along with its optional features that include additional tools for agents, you can do so by using the following command:
```shell
@@ -59,6 +92,22 @@ pip install 'crewai[tools]'
```
The command above installs the basic package and also adds extra components which require more dependencies to function.
### Troubleshooting Dependencies
If you encounter issues during installation or usage, here are some common solutions:
#### Common Issues
1. **ModuleNotFoundError: No module named 'tiktoken'**
- Install tiktoken explicitly: `pip install 'crewai[embeddings]'`
- If using embedchain or other tools: `pip install 'crewai[tools]'`
2. **Failed building wheel for tiktoken**
- Ensure Rust compiler is installed (see installation steps above)
- For Windows: Verify Visual C++ Build Tools are installed
- Try upgrading pip: `pip install --upgrade pip`
- If issues persist, use a pre-built wheel: `pip install tiktoken --prefer-binary`
### 2. Setting Up Your Crew with the YAML Configuration
To create a new CrewAI project, run the following CLI (Command Line Interface) command:
@@ -264,13 +313,16 @@ In addition to the sequential process, you can use the hierarchical process, whi
## Key Features
- **Role-Based Agent Design**: Customize agents with specific roles, goals, and tools.
- **Autonomous Inter-Agent Delegation**: Agents can autonomously delegate tasks and inquire amongst themselves, enhancing problem-solving efficiency.
- **Flexible Task Management**: Define tasks with customizable tools and assign them to agents dynamically.
- **Processes Driven**: Currently only supports `sequential` task execution and `hierarchical` processes, but more complex processes like consensual and autonomous are being worked on.
- **Save output as file**: Save the output of individual tasks as a file, so you can use it later.
- **Parse output as Pydantic or Json**: Parse the output of individual tasks as a Pydantic model or as a Json if you want to.
- **Works with Open Source Models**: Run your crew using Open AI or open source models refer to the [Connect CrewAI to LLMs](https://docs.crewai.com/how-to/LLM-Connections/) page for details on configuring your agents' connections to models, even ones running locally!
**Note**: CrewAI is a standalone framework built from the ground up, without dependencies on Langchain or other agent frameworks.
- **Deep Customization**: Build sophisticated agents with full control over the system - from overriding inner prompts to accessing low-level APIs. Customize roles, goals, tools, and behaviors while maintaining clean abstractions.
- **Autonomous Inter-Agent Delegation**: Agents can autonomously delegate tasks and inquire amongst themselves, enabling complex problem-solving in real-world scenarios.
- **Flexible Task Management**: Define and customize tasks with granular control, from simple operations to complex multi-step processes.
- **Production-Grade Architecture**: Support for both high-level abstractions and low-level customization, with robust error handling and state management.
- **Predictable Results**: Ensure consistent, accurate outputs through programmatic guardrails, agent training capabilities, and flow-based execution control. See our [documentation on guardrails](https://docs.crewai.com/how-to/guardrails/) for implementation details.
- **Model Flexibility**: Run your crew using OpenAI or open source models with production-ready integrations. See [Connect CrewAI to LLMs](https://docs.crewai.com/how-to/LLM-Connections/) for detailed configuration options.
- **Event-Driven Flows**: Build complex, real-world workflows with precise control over execution paths, state management, and conditional logic.
- **Process Orchestration**: Achieve any workflow pattern through flows - from simple sequential and hierarchical processes to complex, custom orchestration patterns with conditional branching and parallel execution.
![CrewAI Mind Map](./docs/crewAI-mindmap.png "CrewAI Mind Map")
@@ -305,6 +357,98 @@ You can test different real life examples of AI crews in the [CrewAI-examples re
[![Stock Analysis](https://img.youtube.com/vi/e0Uj4yWdaAg/maxresdefault.jpg)](https://www.youtube.com/watch?v=e0Uj4yWdaAg "Stock Analysis")
### Using Crews and Flows Together
CrewAI's power truly shines when combining Crews with Flows to create sophisticated automation pipelines. Here's how you can orchestrate multiple Crews within a Flow:
```python
from crewai.flow.flow import Flow, listen, start, router
from crewai import Crew, Agent, Task
from pydantic import BaseModel
# Define structured state for precise control
class MarketState(BaseModel):
sentiment: str = "neutral"
confidence: float = 0.0
recommendations: list = []
class AdvancedAnalysisFlow(Flow[MarketState]):
@start()
def fetch_market_data(self):
# Demonstrate low-level control with structured state
self.state.sentiment = "analyzing"
return {"sector": "tech", "timeframe": "1W"} # These parameters match the task description template
@listen(fetch_market_data)
def analyze_with_crew(self, market_data):
# Show crew agency through specialized roles
analyst = Agent(
role="Senior Market Analyst",
goal="Conduct deep market analysis with expert insight",
backstory="You're a veteran analyst known for identifying subtle market patterns"
)
researcher = Agent(
role="Data Researcher",
goal="Gather and validate supporting market data",
backstory="You excel at finding and correlating multiple data sources"
)
analysis_task = Task(
description="Analyze {sector} sector data for the past {timeframe}",
expected_output="Detailed market analysis with confidence score",
agent=analyst
)
research_task = Task(
description="Find supporting data to validate the analysis",
expected_output="Corroborating evidence and potential contradictions",
agent=researcher
)
# Demonstrate crew autonomy
analysis_crew = Crew(
agents=[analyst, researcher],
tasks=[analysis_task, research_task],
process=Process.sequential,
verbose=True
)
return analysis_crew.kickoff(inputs=market_data) # Pass market_data as named inputs
@router(analyze_with_crew)
def determine_next_steps(self):
# Show flow control with conditional routing
if self.state.confidence > 0.8:
return "high_confidence"
elif self.state.confidence > 0.5:
return "medium_confidence"
return "low_confidence"
@listen("high_confidence")
def execute_strategy(self):
# Demonstrate complex decision making
strategy_crew = Crew(
agents=[
Agent(role="Strategy Expert",
goal="Develop optimal market strategy")
],
tasks=[
Task(description="Create detailed strategy based on analysis",
expected_output="Step-by-step action plan")
]
)
return strategy_crew.kickoff()
@listen("medium_confidence", "low_confidence")
def request_additional_analysis(self):
self.state.recommendations.append("Gather more data")
return "Additional analysis required"
```
This example demonstrates how to:
1. Use Python code for basic data operations
2. Create and execute Crews as steps in your workflow
3. Use Flow decorators to manage the sequence of operations
4. Implement conditional branching based on Crew results
## Connecting Your Crew to a Model
CrewAI supports using various LLMs through a variety of connection options. By default your agents will use the OpenAI API when querying the model. However, there are several other ways to allow your agents to connect to models. For example, you can configure your agents to use a local model via the Ollama tool.
@@ -313,9 +457,13 @@ Please refer to the [Connect CrewAI to LLMs](https://docs.crewai.com/how-to/LLM-
## How CrewAI Compares
**CrewAI's Advantage**: CrewAI is built with production in mind. It offers the flexibility of Autogen's conversational agents and the structured process approach of ChatDev, but without the rigidity. CrewAI's processes are designed to be dynamic and adaptable, fitting seamlessly into both development and production workflows.
**CrewAI's Advantage**: CrewAI combines autonomous agent intelligence with precise workflow control through its unique Crews and Flows architecture. The framework excels at both high-level orchestration and low-level customization, enabling complex, production-grade systems with granular control.
- **Autogen**: While Autogen does good in creating conversational agents capable of working together, it lacks an inherent concept of process. In Autogen, orchestrating agents' interactions requires additional programming, which can become complex and cumbersome as the scale of tasks grows.
- **LangGraph**: While LangGraph provides a foundation for building agent workflows, its approach requires significant boilerplate code and complex state management patterns. The framework's tight coupling with LangChain can limit flexibility when implementing custom agent behaviors or integrating with external systems.
*P.S. CrewAI demonstrates significant performance advantages over LangGraph, executing 5.76x faster in certain cases like this QA task example ([see comparison](https://github.com/crewAIInc/crewAI-examples/tree/main/Notebooks/CrewAI%20Flows%20%26%20Langgraph/QA%20Agent)) while achieving higher evaluation scores with faster completion times in certain coding tasks, like in this example ([detailed analysis](https://github.com/crewAIInc/crewAI-examples/blob/main/Notebooks/CrewAI%20Flows%20%26%20Langgraph/Coding%20Assistant/coding_assistant_eval.ipynb)).*
- **Autogen**: While Autogen excels at creating conversational agents capable of working together, it lacks an inherent concept of process. In Autogen, orchestrating agents' interactions requires additional programming, which can become complex and cumbersome as the scale of tasks grows.
- **ChatDev**: ChatDev introduced the idea of processes into the realm of AI agents, but its implementation is quite rigid. Customizations in ChatDev are limited and not geared towards production environments, which can hinder scalability and flexibility in real-world applications.
@@ -440,5 +588,8 @@ A: CrewAI uses anonymous telemetry to collect usage data for improvement purpose
### Q: Where can I find examples of CrewAI in action?
A: You can find various real-life examples in the [CrewAI-examples repository](https://github.com/crewAIInc/crewAI-examples), including trip planners, stock analysis tools, and more.
### Q: What is the difference between Crews and Flows?
A: Crews and Flows serve different but complementary purposes in CrewAI. Crews are teams of AI agents working together to accomplish specific tasks through role-based collaboration, delivering accurate and predictable results. Flows, on the other hand, are event-driven workflows that can orchestrate both Crews and regular Python code, allowing you to build complex automation pipelines with secure state management and conditional execution paths.
### Q: How can I contribute to CrewAI?
A: Contributions are welcome! You can fork the repository, create a new branch for your feature, add your improvement, and send a pull request. Check the Contribution section in the README for more details.

View File

@@ -32,7 +32,6 @@ A crew in crewAI represents a collaborative group of agents working together to
| **Share Crew** _(optional)_ | `share_crew` | Whether you want to share the complete crew information and execution with the crewAI team to make the library better, and allow us to train models. |
| **Output Log File** _(optional)_ | `output_log_file` | Whether you want to have a file with the complete crew output and execution. You can set it using True and it will default to the folder you are currently in and it will be called logs.txt or passing a string with the full path and name of the file. |
| **Manager Agent** _(optional)_ | `manager_agent` | `manager` sets a custom agent that will be used as a manager. |
| **Manager Callbacks** _(optional)_ | `manager_callbacks` | `manager_callbacks` takes a list of callback handlers to be executed by the manager agent when a hierarchical process is used. |
| **Prompt File** _(optional)_ | `prompt_file` | Path to the prompt JSON file to be used for the crew. |
| **Planning** *(optional)* | `planning` | Adds planning ability to the Crew. When activated before each Crew iteration, all Crew data is sent to an AgentPlanner that will plan the tasks and this plan will be added to each task description. |
| **Planning LLM** *(optional)* | `planning_llm` | The language model used by the AgentPlanner in a planning process. |

View File

@@ -79,6 +79,55 @@ crew = Crew(
result = crew.kickoff(inputs={"question": "What city does John live in and how old is he?"})
```
Here's another example with the `CrewDoclingSource`
```python Code
from crewai import LLM, Agent, Crew, Process, Task
from crewai.knowledge.source.crew_docling_source import CrewDoclingSource
# Create a knowledge source
content_source = CrewDoclingSource(
file_paths=[
"https://lilianweng.github.io/posts/2024-11-28-reward-hacking",
"https://lilianweng.github.io/posts/2024-07-07-hallucination",
],
)
# Create an LLM with a temperature of 0 to ensure deterministic outputs
llm = LLM(model="gpt-4o-mini", temperature=0)
# Create an agent with the knowledge store
agent = Agent(
role="About papers",
goal="You know everything about the papers.",
backstory="""You are a master at understanding papers and their content.""",
verbose=True,
allow_delegation=False,
llm=llm,
)
task = Task(
description="Answer the following questions about the papers: {question}",
expected_output="An answer to the question.",
agent=agent,
)
crew = Crew(
agents=[agent],
tasks=[task],
verbose=True,
process=Process.sequential,
knowledge_sources=[
content_source
], # Enable knowledge by adding the sources here. You can also add more sources to the sources list.
)
result = crew.kickoff(
inputs={
"question": "What is the reward hacking paper about? Be sure to provide sources."
}
)
```
## Knowledge Configuration
### Chunking Configuration
@@ -122,6 +171,58 @@ crewai reset-memories --knowledge
This is useful when you've updated your knowledge sources and want to ensure that the agents are using the most recent information.
## Agent-Specific Knowledge
While knowledge can be provided at the crew level using `crew.knowledge_sources`, individual agents can also have their own knowledge sources using the `knowledge_sources` parameter:
```python Code
from crewai import Agent, Task, Crew
from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource
# Create agent-specific knowledge about a product
product_specs = StringKnowledgeSource(
content="""The XPS 13 laptop features:
- 13.4-inch 4K display
- Intel Core i7 processor
- 16GB RAM
- 512GB SSD storage
- 12-hour battery life""",
metadata={"category": "product_specs"}
)
# Create a support agent with product knowledge
support_agent = Agent(
role="Technical Support Specialist",
goal="Provide accurate product information and support.",
backstory="You are an expert on our laptop products and specifications.",
knowledge_sources=[product_specs] # Agent-specific knowledge
)
# Create a task that requires product knowledge
support_task = Task(
description="Answer this customer question: {question}",
agent=support_agent
)
# Create and run the crew
crew = Crew(
agents=[support_agent],
tasks=[support_task]
)
# Get answer about the laptop's specifications
result = crew.kickoff(
inputs={"question": "What is the storage capacity of the XPS 13?"}
)
```
<Info>
Benefits of agent-specific knowledge:
- Give agents specialized information for their roles
- Maintain separation of concerns between agents
- Combine with crew-level knowledge for layered information access
</Info>
## Custom Knowledge Sources
CrewAI allows you to create custom knowledge sources for any type of data by extending the `BaseKnowledgeSource` class. Let's create a practical example that fetches and processes space news articles.

View File

@@ -29,7 +29,7 @@ Large Language Models (LLMs) are the core intelligence behind CrewAI agents. The
## Available Models and Their Capabilities
Here's a detailed breakdown of supported models and their capabilities:
Here's a detailed breakdown of supported models and their capabilities, you can compare performance at [lmarena.ai](https://lmarena.ai/?leaderboard) and [artificialanalysis.ai](https://artificialanalysis.ai/):
<Tabs>
<Tab title="OpenAI">
@@ -43,13 +43,104 @@ Here's a detailed breakdown of supported models and their capabilities:
1 token ≈ 4 characters in English. For example, 8,192 tokens ≈ 32,768 characters or about 6,000 words.
</Note>
</Tab>
<Tab title="Nvidia NIM">
| Model | Context Window | Best For |
|-------|---------------|-----------|
| nvidia/mistral-nemo-minitron-8b-8k-instruct | 8,192 tokens | State-of-the-art small language model delivering superior accuracy for chatbot, virtual assistants, and content generation. |
| nvidia/nemotron-4-mini-hindi-4b-instruct| 4,096 tokens | A bilingual Hindi-English SLM for on-device inference, tailored specifically for Hindi Language. |
| "nvidia/llama-3.1-nemotron-70b-instruct | 128k tokens | Llama-3.1-Nemotron-70B-Instruct is a large language model customized by NVIDIA in order to improve the helpfulness of LLM generated responses. |
| nvidia/llama3-chatqa-1.5-8b | 128k tokens | Advanced LLM to generate high-quality, context-aware responses for chatbots and search engines. |
| nvidia/llama3-chatqa-1.5-70b | 128k tokens | Advanced LLM to generate high-quality, context-aware responses for chatbots and search engines. |
| nvidia/vila | 128k tokens | Multi-modal vision-language model that understands text/img/video and creates informative responses |
| nvidia/neva-22| 4,096 tokens | Multi-modal vision-language model that understands text/images and generates informative responses |
| nvidia/nemotron-mini-4b-instruct | 8,192 tokens | General-purpose tasks |
| nvidia/usdcode-llama3-70b-instruct | 128k tokens | State-of-the-art LLM that answers OpenUSD knowledge queries and generates USD-Python code. |
| nvidia/nemotron-4-340b-instruct | 4,096 tokens | Creates diverse synthetic data that mimics the characteristics of real-world data. |
| meta/codellama-70b | 100k tokens | LLM capable of generating code from natural language and vice versa. |
| meta/llama2-70b | 4,096 tokens | Cutting-edge large language AI model capable of generating text and code in response to prompts. |
| meta/llama3-8b-instruct | 8,192 tokens | Advanced state-of-the-art LLM with language understanding, superior reasoning, and text generation. |
| meta/llama3-70b-instruct | 8,192 tokens | Powers complex conversations with superior contextual understanding, reasoning and text generation. |
| meta/llama-3.1-8b-instruct | 128k tokens | Advanced state-of-the-art model with language understanding, superior reasoning, and text generation. |
| meta/llama-3.1-70b-instruct | 128k tokens | Powers complex conversations with superior contextual understanding, reasoning and text generation. |
| meta/llama-3.1-405b-instruct | 128k tokens | Advanced LLM for synthetic data generation, distillation, and inference for chatbots, coding, and domain-specific tasks. |
| meta/llama-3.2-1b-instruct | 128k tokens | Advanced state-of-the-art small language model with language understanding, superior reasoning, and text generation. |
| meta/llama-3.2-3b-instruct | 128k tokens | Advanced state-of-the-art small language model with language understanding, superior reasoning, and text generation. |
| meta/llama-3.2-11b-vision-instruct | 128k tokens | Advanced state-of-the-art small language model with language understanding, superior reasoning, and text generation. |
| meta/llama-3.2-90b-vision-instruct | 128k tokens | Advanced state-of-the-art small language model with language understanding, superior reasoning, and text generation. |
| meta/llama-3.1-70b-instruct | 128k tokens | Powers complex conversations with superior contextual understanding, reasoning and text generation. |
| google/gemma-7b | 8,192 tokens | Cutting-edge text generation model text understanding, transformation, and code generation. |
| google/gemma-2b | 8,192 tokens | Cutting-edge text generation model text understanding, transformation, and code generation. |
| google/codegemma-7b | 8,192 tokens | Cutting-edge model built on Google's Gemma-7B specialized for code generation and code completion. |
| google/codegemma-1.1-7b | 8,192 tokens | Advanced programming model for code generation, completion, reasoning, and instruction following. |
| google/recurrentgemma-2b | 8,192 tokens | Novel recurrent architecture based language model for faster inference when generating long sequences. |
| google/gemma-2-9b-it | 8,192 tokens | Cutting-edge text generation model text understanding, transformation, and code generation. |
| google/gemma-2-27b-it | 8,192 tokens | Cutting-edge text generation model text understanding, transformation, and code generation. |
| google/gemma-2-2b-it | 8,192 tokens | Cutting-edge text generation model text understanding, transformation, and code generation. |
| google/deplot | 512 tokens | One-shot visual language understanding model that translates images of plots into tables. |
| google/paligemma | 8,192 tokens | Vision language model adept at comprehending text and visual inputs to produce informative responses. |
| mistralai/mistral-7b-instruct-v0.2 | 32k tokens | This LLM follows instructions, completes requests, and generates creative text. |
| mistralai/mixtral-8x7b-instruct-v0.1 | 8,192 tokens | An MOE LLM that follows instructions, completes requests, and generates creative text. |
| mistralai/mistral-large | 4,096 tokens | Creates diverse synthetic data that mimics the characteristics of real-world data. |
| mistralai/mixtral-8x22b-instruct-v0.1 | 8,192 tokens | Creates diverse synthetic data that mimics the characteristics of real-world data. |
| mistralai/mistral-7b-instruct-v0.3 | 32k tokens | This LLM follows instructions, completes requests, and generates creative text. |
| nv-mistralai/mistral-nemo-12b-instruct | 128k tokens | Most advanced language model for reasoning, code, multilingual tasks; runs on a single GPU. |
| mistralai/mamba-codestral-7b-v0.1 | 256k tokens | Model for writing and interacting with code across a wide range of programming languages and tasks. |
| microsoft/phi-3-mini-128k-instruct | 128K tokens | Lightweight, state-of-the-art open LLM with strong math and logical reasoning skills. |
| microsoft/phi-3-mini-4k-instruct | 4,096 tokens | Lightweight, state-of-the-art open LLM with strong math and logical reasoning skills. |
| microsoft/phi-3-small-8k-instruct | 8,192 tokens | Lightweight, state-of-the-art open LLM with strong math and logical reasoning skills. |
| microsoft/phi-3-small-128k-instruct | 128K tokens | Lightweight, state-of-the-art open LLM with strong math and logical reasoning skills. |
| microsoft/phi-3-medium-4k-instruct | 4,096 tokens | Lightweight, state-of-the-art open LLM with strong math and logical reasoning skills. |
| microsoft/phi-3-medium-128k-instruct | 128K tokens | Lightweight, state-of-the-art open LLM with strong math and logical reasoning skills. |
| microsoft/phi-3.5-mini-instruct | 128K tokens | Lightweight multilingual LLM powering AI applications in latency bound, memory/compute constrained environments |
| microsoft/phi-3.5-moe-instruct | 128K tokens | Advanced LLM based on Mixture of Experts architecure to deliver compute efficient content generation |
| microsoft/kosmos-2 | 1,024 tokens | Groundbreaking multimodal model designed to understand and reason about visual elements in images. |
| microsoft/phi-3-vision-128k-instruct | 128k tokens | Cutting-edge open multimodal model exceling in high-quality reasoning from images. |
| microsoft/phi-3.5-vision-instruct | 128k tokens | Cutting-edge open multimodal model exceling in high-quality reasoning from images. |
| databricks/dbrx-instruct | 12k tokens | A general-purpose LLM with state-of-the-art performance in language understanding, coding, and RAG. |
| snowflake/arctic | 1,024 tokens | Delivers high efficiency inference for enterprise applications focused on SQL generation and coding. |
| aisingapore/sea-lion-7b-instruct | 4,096 tokens | LLM to represent and serve the linguistic and cultural diversity of Southeast Asia |
| ibm/granite-8b-code-instruct | 4,096 tokens | Software programming LLM for code generation, completion, explanation, and multi-turn conversion. |
| ibm/granite-34b-code-instruct | 8,192 tokens | Software programming LLM for code generation, completion, explanation, and multi-turn conversion. |
| ibm/granite-3.0-8b-instruct | 4,096 tokens | Advanced Small Language Model supporting RAG, summarization, classification, code, and agentic AI |
| ibm/granite-3.0-3b-a800m-instruct | 4,096 tokens | Highly efficient Mixture of Experts model for RAG, summarization, entity extraction, and classification |
| mediatek/breeze-7b-instruct | 4,096 tokens | Creates diverse synthetic data that mimics the characteristics of real-world data. |
| upstage/solar-10.7b-instruct | 4,096 tokens | Excels in NLP tasks, particularly in instruction-following, reasoning, and mathematics. |
| writer/palmyra-med-70b-32k | 32k tokens | Leading LLM for accurate, contextually relevant responses in the medical domain. |
| writer/palmyra-med-70b | 32k tokens | Leading LLM for accurate, contextually relevant responses in the medical domain. |
| writer/palmyra-fin-70b-32k | 32k tokens | Specialized LLM for financial analysis, reporting, and data processing |
| 01-ai/yi-large | 32k tokens | Powerful model trained on English and Chinese for diverse tasks including chatbot and creative writing. |
| deepseek-ai/deepseek-coder-6.7b-instruct | 2k tokens | Powerful coding model offering advanced capabilities in code generation, completion, and infilling |
| rakuten/rakutenai-7b-instruct | 1,024 tokens | Advanced state-of-the-art LLM with language understanding, superior reasoning, and text generation. |
| rakuten/rakutenai-7b-chat | 1,024 tokens | Advanced state-of-the-art LLM with language understanding, superior reasoning, and text generation. |
| baichuan-inc/baichuan2-13b-chat | 4,096 tokens | Support Chinese and English chat, coding, math, instruction following, solving quizzes |
<Note>
NVIDIA's NIM support for models is expanding continuously! For the most up-to-date list of available models, please visit build.nvidia.com.
</Note>
</Tab>
<Tab title="Gemini">
| Model | Context Window | Best For |
|-------|---------------|-----------|
| gemini-2.0-flash-exp | 1M tokens | Higher quality at faster speed, multimodal model, good for most tasks |
| gemini-1.5-flash | 1M tokens | Balanced multimodal model, good for most tasks |
| gemini-1.5-flash-8B | 1M tokens | Fastest, most cost-efficient, good for high-frequency tasks |
| gemini-1.5-pro | 2M tokens | Best performing, wide variety of reasoning tasks including logical reasoning, coding, and creative collaboration |
<Tip>
Google's Gemini models are all multimodal, supporting audio, images, video and text, supporting context caching, json schema, function calling, etc.
These models are available via API_KEY from
[The Gemini API](https://ai.google.dev/gemini-api/docs) and also from
[Google Cloud Vertex](https://cloud.google.com/vertex-ai/generative-ai/docs/migrate/migrate-google-ai) as part of the
[Model Garden](https://cloud.google.com/vertex-ai/generative-ai/docs/model-garden/explore-models).
</Tip>
</Tab>
<Tab title="Groq">
| Model | Context Window | Best For |
|-------|---------------|-----------|
| Llama 3.1 70B/8B | 131,072 tokens | High-performance, large context tasks |
| Llama 3.2 Series | 8,192 tokens | General-purpose tasks |
| Mixtral 8x7B | 32,768 tokens | Balanced performance and context |
| Gemma Series | 8,192 tokens | Efficient, smaller-scale tasks |
<Tip>
Groq is known for its fast inference speeds, making it suitable for real-time applications.
@@ -60,7 +151,7 @@ Here's a detailed breakdown of supported models and their capabilities:
|----------|---------------|--------------|
| Deepseek Chat | 128,000 tokens | Specialized in technical discussions |
| Claude 3 | Up to 200K tokens | Strong reasoning, code understanding |
| Gemini | Varies by model | Multimodal capabilities |
| Gemma Series | 8,192 tokens | Efficient, smaller-scale tasks |
<Info>
Provider selection should consider factors like:
@@ -128,10 +219,10 @@ There are three ways to configure LLMs in CrewAI. Choose the method that best fi
# llm: anthropic/claude-2.1
# llm: anthropic/claude-2.0
# Google Models - Good for general tasks
# llm: gemini/gemini-pro
# Google Models - Strong reasoning, large cachable context window, multimodal
# llm: gemini/gemini-1.5-pro-latest
# llm: gemini/gemini-1.0-pro-latest
# llm: gemini/gemini-1.5-flash-latest
# llm: gemini/gemini-1.5-flash-8b-latest
# AWS Bedrock Models - Enterprise-grade
# llm: bedrock/anthropic.claude-3-sonnet-20240229-v1:0
@@ -350,13 +441,18 @@ Learn how to get the most out of your LLM configuration:
<Accordion title="Google">
```python Code
# Option 1. Gemini accessed with an API key.
# https://ai.google.dev/gemini-api/docs/api-key
GEMINI_API_KEY=<your-api-key>
# Option 2. Vertex AI IAM credentials for Gemini, Anthropic, and anything in the Model Garden.
# https://cloud.google.com/vertex-ai/generative-ai/docs/overview
```
Example usage:
```python Code
llm = LLM(
model="gemini/gemini-pro",
model="gemini/gemini-1.5-pro-latest",
temperature=0.7
)
```
@@ -412,6 +508,20 @@ Learn how to get the most out of your LLM configuration:
```
</Accordion>
<Accordion title="Nvidia NIM">
```python Code
NVIDIA_API_KEY=<your-api-key>
```
Example usage:
```python Code
llm = LLM(
model="nvidia_nim/meta/llama3-70b-instruct",
temperature=0.7
)
```
</Accordion>
<Accordion title="Groq">
```python Code
GROQ_API_KEY=<your-api-key>
@@ -502,20 +612,6 @@ Learn how to get the most out of your LLM configuration:
```
</Accordion>
<Accordion title="Nvidia NIM">
```python Code
NVIDIA_API_KEY=<your-api-key>
```
Example usage:
```python Code
llm = LLM(
model="nvidia_nim/meta/llama3-70b-instruct",
temperature=0.7
)
```
</Accordion>
<Accordion title="SambaNova">
```python Code
SAMBANOVA_API_KEY=<your-api-key>

View File

@@ -6,7 +6,7 @@ icon: list-check
## Overview of a Task
In the CrewAI framework, a `Task` is a specific assignment completed by an `Agent`.
In the CrewAI framework, a `Task` is a specific assignment completed by an `Agent`.
Tasks provide all necessary details for execution, such as a description, the agent responsible, required tools, and more, facilitating a wide range of action complexities.
@@ -263,8 +263,148 @@ analysis_task = Task(
)
```
## Task Guardrails
Task guardrails provide a way to validate and transform task outputs before they
are passed to the next task. This feature helps ensure data quality and provides
efeedback to agents when their output doesn't meet specific criteria.
### Using Task Guardrails
To add a guardrail to a task, provide a validation function through the `guardrail` parameter:
```python Code
from typing import Tuple, Union, Dict, Any
def validate_blog_content(result: str) -> Tuple[bool, Union[Dict[str, Any], str]]:
"""Validate blog content meets requirements."""
try:
# Check word count
word_count = len(result.split())
if word_count > 200:
return (False, {
"error": "Blog content exceeds 200 words",
"code": "WORD_COUNT_ERROR",
"context": {"word_count": word_count}
})
# Additional validation logic here
return (True, result.strip())
except Exception as e:
return (False, {
"error": "Unexpected error during validation",
"code": "SYSTEM_ERROR"
})
blog_task = Task(
description="Write a blog post about AI",
expected_output="A blog post under 200 words",
agent=blog_agent,
guardrail=validate_blog_content # Add the guardrail function
)
```
### Guardrail Function Requirements
1. **Function Signature**:
- Must accept exactly one parameter (the task output)
- Should return a tuple of `(bool, Any)`
- Type hints are recommended but optional
2. **Return Values**:
- Success: Return `(True, validated_result)`
- Failure: Return `(False, error_details)`
### Error Handling Best Practices
1. **Structured Error Responses**:
```python Code
def validate_with_context(result: str) -> Tuple[bool, Union[Dict[str, Any], str]]:
try:
# Main validation logic
validated_data = perform_validation(result)
return (True, validated_data)
except ValidationError as e:
return (False, {
"error": str(e),
"code": "VALIDATION_ERROR",
"context": {"input": result}
})
except Exception as e:
return (False, {
"error": "Unexpected error",
"code": "SYSTEM_ERROR"
})
```
2. **Error Categories**:
- Use specific error codes
- Include relevant context
- Provide actionable feedback
3. **Validation Chain**:
```python Code
from typing import Any, Dict, List, Tuple, Union
def complex_validation(result: str) -> Tuple[bool, Union[str, Dict[str, Any]]]:
"""Chain multiple validation steps."""
# Step 1: Basic validation
if not result:
return (False, {"error": "Empty result", "code": "EMPTY_INPUT"})
# Step 2: Content validation
try:
validated = validate_content(result)
if not validated:
return (False, {"error": "Invalid content", "code": "CONTENT_ERROR"})
# Step 3: Format validation
formatted = format_output(validated)
return (True, formatted)
except Exception as e:
return (False, {
"error": str(e),
"code": "VALIDATION_ERROR",
"context": {"step": "content_validation"}
})
```
### Handling Guardrail Results
When a guardrail returns `(False, error)`:
1. The error is sent back to the agent
2. The agent attempts to fix the issue
3. The process repeats until:
- The guardrail returns `(True, result)`
- Maximum retries are reached
Example with retry handling:
```python Code
from typing import Optional, Tuple, Union
def validate_json_output(result: str) -> Tuple[bool, Union[Dict[str, Any], str]]:
"""Validate and parse JSON output."""
try:
# Try to parse as JSON
data = json.loads(result)
return (True, data)
except json.JSONDecodeError as e:
return (False, {
"error": "Invalid JSON format",
"code": "JSON_ERROR",
"context": {"line": e.lineno, "column": e.colno}
})
task = Task(
description="Generate a JSON report",
expected_output="A valid JSON object",
agent=analyst,
guardrail=validate_json_output,
max_retries=3 # Limit retry attempts
)
```
## Getting Structured Consistent Outputs from Tasks
When you need to ensure that a task outputs a structured and consistent format, you can use the `output_pydantic` or `output_json` properties on a task. These properties allow you to define the expected output structure, making it easier to parse and utilize the results in your application.
<Note>
It's also important to note that the output of the final task of a crew becomes the final output of the actual crew itself.
@@ -608,6 +748,114 @@ While creating and executing tasks, certain validation mechanisms are in place t
These validations help in maintaining the consistency and reliability of task executions within the crewAI framework.
## Task Guardrails
Task guardrails provide a powerful way to validate, transform, or filter task outputs before they are passed to the next task. Guardrails are optional functions that execute before the next task starts, allowing you to ensure that task outputs meet specific requirements or formats.
### Basic Usage
```python Code
from typing import Tuple, Union
from crewai import Task
def validate_json_output(result: str) -> Tuple[bool, Union[dict, str]]:
"""Validate that the output is valid JSON."""
try:
json_data = json.loads(result)
return (True, json_data)
except json.JSONDecodeError:
return (False, "Output must be valid JSON")
task = Task(
description="Generate JSON data",
expected_output="Valid JSON object",
guardrail=validate_json_output
)
```
### How Guardrails Work
1. **Optional Attribute**: Guardrails are an optional attribute at the task level, allowing you to add validation only where needed.
2. **Execution Timing**: The guardrail function is executed before the next task starts, ensuring valid data flow between tasks.
3. **Return Format**: Guardrails must return a tuple of `(success, data)`:
- If `success` is `True`, `data` is the validated/transformed result
- If `success` is `False`, `data` is the error message
4. **Result Routing**:
- On success (`True`), the result is automatically passed to the next task
- On failure (`False`), the error is sent back to the agent to generate a new answer
### Common Use Cases
#### Data Format Validation
```python Code
def validate_email_format(result: str) -> Tuple[bool, Union[str, str]]:
"""Ensure the output contains a valid email address."""
import re
email_pattern = r'^[\w\.-]+@[\w\.-]+\.\w+$'
if re.match(email_pattern, result.strip()):
return (True, result.strip())
return (False, "Output must be a valid email address")
```
#### Content Filtering
```python Code
def filter_sensitive_info(result: str) -> Tuple[bool, Union[str, str]]:
"""Remove or validate sensitive information."""
sensitive_patterns = ['SSN:', 'password:', 'secret:']
for pattern in sensitive_patterns:
if pattern.lower() in result.lower():
return (False, f"Output contains sensitive information ({pattern})")
return (True, result)
```
#### Data Transformation
```python Code
def normalize_phone_number(result: str) -> Tuple[bool, Union[str, str]]:
"""Ensure phone numbers are in a consistent format."""
import re
digits = re.sub(r'\D', '', result)
if len(digits) == 10:
formatted = f"({digits[:3]}) {digits[3:6]}-{digits[6:]}"
return (True, formatted)
return (False, "Output must be a 10-digit phone number")
```
### Advanced Features
#### Chaining Multiple Validations
```python Code
def chain_validations(*validators):
"""Chain multiple validators together."""
def combined_validator(result):
for validator in validators:
success, data = validator(result)
if not success:
return (False, data)
result = data
return (True, result)
return combined_validator
# Usage
task = Task(
description="Get user contact info",
expected_output="Email and phone",
guardrail=chain_validations(
validate_email_format,
filter_sensitive_info
)
)
```
#### Custom Retry Logic
```python Code
task = Task(
description="Generate data",
expected_output="Valid data",
guardrail=validate_data,
max_retries=5 # Override default retry limit
)
```
## Creating Directories when Saving Files
You can now specify if a task should create directories when saving its output to a file. This is particularly useful for organizing outputs and ensuring that file paths are correctly structured.
@@ -629,7 +877,7 @@ save_output_task = Task(
## Conclusion
Tasks are the driving force behind the actions of agents in CrewAI.
By properly defining tasks and their outcomes, you set the stage for your AI agents to work effectively, either independently or as a collaborative unit.
Equipping tasks with appropriate tools, understanding the execution process, and following robust validation practices are crucial for maximizing CrewAI's potential,
Tasks are the driving force behind the actions of agents in CrewAI.
By properly defining tasks and their outcomes, you set the stage for your AI agents to work effectively, either independently or as a collaborative unit.
Equipping tasks with appropriate tools, understanding the execution process, and following robust validation practices are crucial for maximizing CrewAI's potential,
ensuring agents are effectively prepared for their assignments and that tasks are executed as intended.

View File

@@ -0,0 +1,211 @@
# Portkey Integration with CrewAI
<img src="https://raw.githubusercontent.com/siddharthsambharia-portkey/Portkey-Product-Images/main/Portkey-CrewAI.png" alt="Portkey CrewAI Header Image" width="70%" />
[Portkey](https://portkey.ai/?utm_source=crewai&utm_medium=crewai&utm_campaign=crewai) is a 2-line upgrade to make your CrewAI agents reliable, cost-efficient, and fast.
Portkey adds 4 core production capabilities to any CrewAI agent:
1. Routing to **200+ LLMs**
2. Making each LLM call more robust
3. Full-stack tracing & cost, performance analytics
4. Real-time guardrails to enforce behavior
## Getting Started
1. **Install Required Packages:**
```bash
pip install -qU crewai portkey-ai
```
2. **Configure the LLM Client:**
To build CrewAI Agents with Portkey, you'll need two keys:
- **Portkey API Key**: Sign up on the [Portkey app](https://app.portkey.ai/?utm_source=crewai&utm_medium=crewai&utm_campaign=crewai) and copy your API key
- **Virtual Key**: Virtual Keys securely manage your LLM API keys in one place. Store your LLM provider API keys securely in Portkey's vault
```python
from crewai import LLM
from portkey_ai import createHeaders, PORTKEY_GATEWAY_URL
gpt_llm = LLM(
model="gpt-4",
base_url=PORTKEY_GATEWAY_URL,
api_key="dummy", # We are using Virtual key
extra_headers=createHeaders(
api_key="YOUR_PORTKEY_API_KEY",
virtual_key="YOUR_VIRTUAL_KEY", # Enter your Virtual key from Portkey
)
)
```
3. **Create and Run Your First Agent:**
```python
from crewai import Agent, Task, Crew
# Define your agents with roles and goals
coder = Agent(
role='Software developer',
goal='Write clear, concise code on demand',
backstory='An expert coder with a keen eye for software trends.',
llm=gpt_llm
)
# Create tasks for your agents
task1 = Task(
description="Define the HTML for making a simple website with heading- Hello World! Portkey is working!",
expected_output="A clear and concise HTML code",
agent=coder
)
# Instantiate your crew
crew = Crew(
agents=[coder],
tasks=[task1],
)
result = crew.kickoff()
print(result)
```
## Key Features
| Feature | Description |
|---------|-------------|
| 🌐 Multi-LLM Support | Access OpenAI, Anthropic, Gemini, Azure, and 250+ providers through a unified interface |
| 🛡️ Production Reliability | Implement retries, timeouts, load balancing, and fallbacks |
| 📊 Advanced Observability | Track 40+ metrics including costs, tokens, latency, and custom metadata |
| 🔍 Comprehensive Logging | Debug with detailed execution traces and function call logs |
| 🚧 Security Controls | Set budget limits and implement role-based access control |
| 🔄 Performance Analytics | Capture and analyze feedback for continuous improvement |
| 💾 Intelligent Caching | Reduce costs and latency with semantic or simple caching |
## Production Features with Portkey Configs
All features mentioned below are through Portkey's Config system. Portkey's Config system allows you to define routing strategies using simple JSON objects in your LLM API calls. You can create and manage Configs directly in your code or through the Portkey Dashboard. Each Config has a unique ID for easy reference.
<Frame>
<img src="https://raw.githubusercontent.com/Portkey-AI/docs-core/refs/heads/main/images/libraries/libraries-3.avif"/>
</Frame>
### 1. Use 250+ LLMs
Access various LLMs like Anthropic, Gemini, Mistral, Azure OpenAI, and more with minimal code changes. Switch between providers or use them together seamlessly. [Learn more about Universal API](https://portkey.ai/docs/product/ai-gateway/universal-api)
Easily switch between different LLM providers:
```python
# Anthropic Configuration
anthropic_llm = LLM(
model="claude-3-5-sonnet-latest",
base_url=PORTKEY_GATEWAY_URL,
api_key="dummy",
extra_headers=createHeaders(
api_key="YOUR_PORTKEY_API_KEY",
virtual_key="YOUR_ANTHROPIC_VIRTUAL_KEY", #You don't need provider when using Virtual keys
trace_id="anthropic_agent"
)
)
# Azure OpenAI Configuration
azure_llm = LLM(
model="gpt-4",
base_url=PORTKEY_GATEWAY_URL,
api_key="dummy",
extra_headers=createHeaders(
api_key="YOUR_PORTKEY_API_KEY",
virtual_key="YOUR_AZURE_VIRTUAL_KEY", #You don't need provider when using Virtual keys
trace_id="azure_agent"
)
)
```
### 2. Caching
Improve response times and reduce costs with two powerful caching modes:
- **Simple Cache**: Perfect for exact matches
- **Semantic Cache**: Matches responses for requests that are semantically similar
[Learn more about Caching](https://portkey.ai/docs/product/ai-gateway/cache-simple-and-semantic)
```py
config = {
"cache": {
"mode": "semantic", # or "simple" for exact matching
}
}
```
### 3. Production Reliability
Portkey provides comprehensive reliability features:
- **Automatic Retries**: Handle temporary failures gracefully
- **Request Timeouts**: Prevent hanging operations
- **Conditional Routing**: Route requests based on specific conditions
- **Fallbacks**: Set up automatic provider failovers
- **Load Balancing**: Distribute requests efficiently
[Learn more about Reliability Features](https://portkey.ai/docs/product/ai-gateway/)
### 4. Metrics
Agent runs are complex. Portkey automatically logs **40+ comprehensive metrics** for your AI agents, including cost, tokens used, latency, etc. Whether you need a broad overview or granular insights into your agent runs, Portkey's customizable filters provide the metrics you need.
- Cost per agent interaction
- Response times and latency
- Token usage and efficiency
- Success/failure rates
- Cache hit rates
<img src="https://github.com/siddharthsambharia-portkey/Portkey-Product-Images/blob/main/Portkey-Dashboard.png?raw=true" width="70%" alt="Portkey Dashboard" />
### 5. Detailed Logging
Logs are essential for understanding agent behavior, diagnosing issues, and improving performance. They provide a detailed record of agent activities and tool use, which is crucial for debugging and optimizing processes.
Access a dedicated section to view records of agent executions, including parameters, outcomes, function calls, and errors. Filter logs based on multiple parameters such as trace ID, model, tokens used, and metadata.
<details>
<summary><b>Traces</b></summary>
<img src="https://raw.githubusercontent.com/siddharthsambharia-portkey/Portkey-Product-Images/main/Portkey-Traces.png" alt="Portkey Traces" width="70%" />
</details>
<details>
<summary><b>Logs</b></summary>
<img src="https://raw.githubusercontent.com/siddharthsambharia-portkey/Portkey-Product-Images/main/Portkey-Logs.png" alt="Portkey Logs" width="70%" />
</details>
### 6. Enterprise Security Features
- Set budget limit and rate limts per Virtual Key (disposable API keys)
- Implement role-based access control
- Track system changes with audit logs
- Configure data retention policies
For detailed information on creating and managing Configs, visit the [Portkey documentation](https://docs.portkey.ai/product/ai-gateway/configs).
## Resources
- [📘 Portkey Documentation](https://docs.portkey.ai)
- [📊 Portkey Dashboard](https://app.portkey.ai/?utm_source=crewai&utm_medium=crewai&utm_campaign=crewai)
- [🐦 Twitter](https://twitter.com/portkeyai)
- [💬 Discord Community](https://discord.gg/DD7vgKK299)

View File

@@ -0,0 +1,138 @@
---
title: Using Multimodal Agents
description: Learn how to enable and use multimodal capabilities in your agents for processing images and other non-text content within the CrewAI framework.
icon: image
---
# Using Multimodal Agents
CrewAI supports multimodal agents that can process both text and non-text content like images. This guide will show you how to enable and use multimodal capabilities in your agents.
## Enabling Multimodal Capabilities
To create a multimodal agent, simply set the `multimodal` parameter to `True` when initializing your agent:
```python
from crewai import Agent
agent = Agent(
role="Image Analyst",
goal="Analyze and extract insights from images",
backstory="An expert in visual content interpretation with years of experience in image analysis",
multimodal=True # This enables multimodal capabilities
)
```
When you set `multimodal=True`, the agent is automatically configured with the necessary tools for handling non-text content, including the `AddImageTool`.
## Working with Images
The multimodal agent comes pre-configured with the `AddImageTool`, which allows it to process images. You don't need to manually add this tool - it's automatically included when you enable multimodal capabilities.
Here's a complete example showing how to use a multimodal agent to analyze an image:
```python
from crewai import Agent, Task, Crew
# Create a multimodal agent
image_analyst = Agent(
role="Product Analyst",
goal="Analyze product images and provide detailed descriptions",
backstory="Expert in visual product analysis with deep knowledge of design and features",
multimodal=True
)
# Create a task for image analysis
task = Task(
description="Analyze the product image at https://example.com/product.jpg and provide a detailed description",
agent=image_analyst
)
# Create and run the crew
crew = Crew(
agents=[image_analyst],
tasks=[task]
)
result = crew.kickoff()
```
### Advanced Usage with Context
You can provide additional context or specific questions about the image when creating tasks for multimodal agents. The task description can include specific aspects you want the agent to focus on:
```python
from crewai import Agent, Task, Crew
# Create a multimodal agent for detailed analysis
expert_analyst = Agent(
role="Visual Quality Inspector",
goal="Perform detailed quality analysis of product images",
backstory="Senior quality control expert with expertise in visual inspection",
multimodal=True # AddImageTool is automatically included
)
# Create a task with specific analysis requirements
inspection_task = Task(
description="""
Analyze the product image at https://example.com/product.jpg with focus on:
1. Quality of materials
2. Manufacturing defects
3. Compliance with standards
Provide a detailed report highlighting any issues found.
""",
agent=expert_analyst
)
# Create and run the crew
crew = Crew(
agents=[expert_analyst],
tasks=[inspection_task]
)
result = crew.kickoff()
```
### Tool Details
When working with multimodal agents, the `AddImageTool` is automatically configured with the following schema:
```python
class AddImageToolSchema:
image_url: str # Required: The URL or path of the image to process
action: Optional[str] = None # Optional: Additional context or specific questions about the image
```
The multimodal agent will automatically handle the image processing through its built-in tools, allowing it to:
- Access images via URLs or local file paths
- Process image content with optional context or specific questions
- Provide analysis and insights based on the visual information and task requirements
## Best Practices
When working with multimodal agents, keep these best practices in mind:
1. **Image Access**
- Ensure your images are accessible via URLs that the agent can reach
- For local images, consider hosting them temporarily or using absolute file paths
- Verify that image URLs are valid and accessible before running tasks
2. **Task Description**
- Be specific about what aspects of the image you want the agent to analyze
- Include clear questions or requirements in the task description
- Consider using the optional `action` parameter for focused analysis
3. **Resource Management**
- Image processing may require more computational resources than text-only tasks
- Some language models may require base64 encoding for image data
- Consider batch processing for multiple images to optimize performance
4. **Environment Setup**
- Verify that your environment has the necessary dependencies for image processing
- Ensure your language model supports multimodal capabilities
- Test with small images first to validate your setup
5. **Error Handling**
- Implement proper error handling for image loading failures
- Have fallback strategies for when image processing fails
- Monitor and log image processing operations for debugging

View File

@@ -7,7 +7,7 @@ icon: wrench
<Note>
**Python Version Requirements**
CrewAI requires `Python >=3.10 and <=3.12`. Here's how to check your version:
CrewAI requires `Python >=3.10 and <3.13`. Here's how to check your version:
```bash
python3 --version
```

View File

@@ -0,0 +1,222 @@
---
title: BraveSearchTool
description: A tool for performing web searches using the Brave Search API
icon: search
---
## BraveSearchTool
The BraveSearchTool enables web searches using the Brave Search API, providing customizable result counts, country-specific searches, and rate-limited operations. It formats search results with titles, URLs, and snippets for easy consumption.
## Installation
```bash
pip install 'crewai[tools]'
```
## Authentication
Set up your Brave Search API key:
```bash
export BRAVE_API_KEY='your-brave-api-key'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import BraveSearchTool
# Basic initialization
search_tool = BraveSearchTool()
# Advanced initialization with custom parameters
search_tool = BraveSearchTool(
country="US", # Country-specific search
n_results=5, # Number of results to return
save_file=True # Save results to file
)
# Create an agent with the tool
researcher = Agent(
role='Web Researcher',
goal='Search and analyze web content',
backstory='Expert at finding relevant information online.',
tools=[search_tool],
verbose=True
)
```
## Input Schema
```python
class BraveSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the internet"
)
```
## Function Signature
```python
def __init__(
self,
country: Optional[str] = "",
n_results: int = 10,
save_file: bool = False,
*args,
**kwargs
):
"""
Initialize the Brave search tool.
Args:
country (Optional[str]): Country code for region-specific search
n_results (int): Number of results to return (default: 10)
save_file (bool): Whether to save results to file (default: False)
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Execute web search using Brave Search API.
Args:
search_query (str): Query to search
save_file (bool, optional): Override save_file setting
n_results (int, optional): Override n_results setting
Returns:
str: Formatted search results with titles, URLs, and snippets
"""
```
## Best Practices
1. API Authentication:
- Securely store BRAVE_API_KEY
- Keep API key confidential
- Handle authentication errors
2. Rate Limiting:
- Tool automatically handles rate limiting
- Minimum 1-second interval between requests
- Consider implementing additional rate limits
3. Search Optimization:
- Use specific search queries
- Adjust result count based on needs
- Consider regional search requirements
4. Error Handling:
- Handle API request failures
- Manage parsing errors
- Monitor rate limit errors
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import BraveSearchTool
# Initialize tool with custom configuration
search_tool = BraveSearchTool(
country="GB", # UK-specific search
n_results=3, # Limit to 3 results
save_file=True # Save results to file
)
# Create agent
researcher = Agent(
role='Web Researcher',
goal='Research latest AI developments',
backstory='Expert at finding and analyzing tech news.',
tools=[search_tool]
)
# Define task
research_task = Task(
description="""Find the latest news about artificial
intelligence developments in quantum computing.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "latest quantum computing AI developments"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Country-Specific Search
```python
# Initialize tools for different regions
us_search = BraveSearchTool(country="US")
uk_search = BraveSearchTool(country="GB")
jp_search = BraveSearchTool(country="JP")
# Compare results across regions
us_results = us_search.run(
search_query="local news"
)
uk_results = uk_search.run(
search_query="local news"
)
jp_results = jp_search.run(
search_query="local news"
)
```
### Result Management
```python
# Save results to file
archival_search = BraveSearchTool(
save_file=True,
n_results=20
)
# Search and save
results = archival_search.run(
search_query="historical events 2023"
)
# Results saved to search_results_YYYY-MM-DD_HH-MM-SS.txt
```
### Error Handling Example
```python
try:
search_tool = BraveSearchTool()
results = search_tool.run(
search_query="important topic"
)
print(results)
except ValueError as e: # API key missing
print(f"Authentication error: {str(e)}")
except Exception as e:
print(f"Search error: {str(e)}")
```
## Notes
- Requires Brave Search API key
- Implements automatic rate limiting
- Supports country-specific searches
- Customizable result count
- Optional file saving feature
- Thread-safe operations
- Efficient result formatting
- Handles API errors gracefully
- Supports parallel searches
- Maintains search context

View File

@@ -0,0 +1,164 @@
---
title: CodeDocsSearchTool
description: A semantic search tool for code documentation websites using RAG capabilities
icon: book-open
---
## CodeDocsSearchTool
The CodeDocsSearchTool is a specialized Retrieval-Augmented Generation (RAG) tool that enables semantic search within code documentation websites. It inherits from the base RagTool class and provides both fixed and dynamic documentation URL searching capabilities.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import CodeDocsSearchTool
# Method 1: Dynamic documentation URL
docs_search = CodeDocsSearchTool()
# Method 2: Fixed documentation URL
fixed_docs_search = CodeDocsSearchTool(
docs_url="https://docs.example.com"
)
# Create an agent with the tool
researcher = Agent(
role='Documentation Researcher',
goal='Search through code documentation semantically',
backstory='Expert at finding relevant information in technical documentation.',
tools=[docs_search],
verbose=True
)
```
## Input Schema
The tool supports two input schemas depending on initialization:
### Dynamic URL Schema
```python
class CodeDocsSearchToolSchema(BaseModel):
search_query: str # The semantic search query
docs_url: str # URL of the documentation site to search
```
### Fixed URL Schema
```python
class FixedCodeDocsSearchToolSchema(BaseModel):
search_query: str # The semantic search query
```
## Function Signature
```python
def __init__(self, docs_url: Optional[str] = None, **kwargs):
"""
Initialize the documentation search tool.
Args:
docs_url (Optional[str]): Fixed URL to a documentation site. If provided,
the tool will only search this documentation.
**kwargs: Additional arguments passed to the parent RagTool
"""
def _run(self, search_query: str, **kwargs: Any) -> Any:
"""
Perform semantic search on the documentation site.
Args:
search_query (str): The semantic search query
**kwargs: Additional arguments (including 'docs_url' for dynamic mode)
Returns:
str: Relevant documentation passages based on semantic search
"""
```
## Best Practices
1. Choose initialization method based on use case:
- Use fixed URL when repeatedly searching the same documentation
- Use dynamic URL when searching different documentation sites
2. Write clear, semantic search queries
3. Ensure documentation sites are accessible
4. Consider documentation structure and size
5. Handle potential URL access errors in agent prompts
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import CodeDocsSearchTool
# Example 1: Fixed documentation search
api_docs_search = CodeDocsSearchTool(
docs_url="https://api.example.com/docs"
)
# Example 2: Dynamic documentation search
flexible_docs_search = CodeDocsSearchTool()
# Create agents
api_analyst = Agent(
role='API Documentation Analyst',
goal='Find relevant API endpoints and usage examples',
backstory='Expert at analyzing API documentation.',
tools=[api_docs_search]
)
docs_researcher = Agent(
role='Documentation Researcher',
goal='Search through various documentation sites',
backstory='Specialist in finding information across multiple docs.',
tools=[flexible_docs_search]
)
# Define tasks
fixed_search_task = Task(
description="""Find all authentication-related endpoints
in the API documentation.""",
agent=api_analyst
)
# The agent will use:
# {
# "search_query": "authentication endpoints and methods"
# }
dynamic_search_task = Task(
description="""Search through the Python documentation at
docs.python.org for information about async/await.""",
agent=docs_researcher
)
# The agent will use:
# {
# "search_query": "async await syntax and usage",
# "docs_url": "https://docs.python.org"
# }
# Create crew
crew = Crew(
agents=[api_analyst, docs_researcher],
tasks=[fixed_search_task, dynamic_search_task]
)
# Execute
result = crew.kickoff()
```
## Notes
- Inherits from RagTool for semantic search capabilities
- Supports both fixed and dynamic documentation URLs
- Uses embeddings for semantic search
- Thread-safe operations
- Automatically handles documentation loading and embedding
- Optimized for technical documentation search

View File

@@ -0,0 +1,224 @@
---
title: CodeInterpreterTool
description: A tool for secure Python code execution in isolated Docker environments
icon: code
---
## CodeInterpreterTool
The CodeInterpreterTool provides secure Python code execution capabilities using Docker containers. It supports dynamic library installation and offers both safe (Docker-based) and unsafe (direct) execution modes.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import CodeInterpreterTool
# Initialize the tool
code_tool = CodeInterpreterTool()
# Create an agent with the tool
programmer = Agent(
role='Code Executor',
goal='Execute and analyze Python code',
backstory='Expert at writing and executing Python code.',
tools=[code_tool],
verbose=True
)
```
## Input Schema
```python
class CodeInterpreterSchema(BaseModel):
code: str = Field(
description="Python3 code used to be interpreted in the Docker container. ALWAYS PRINT the final result and the output of the code"
)
libraries_used: List[str] = Field(
description="List of libraries used in the code with proper installing names separated by commas. Example: numpy,pandas,beautifulsoup4"
)
```
## Function Signature
```python
def __init__(
self,
code: Optional[str] = None,
user_dockerfile_path: Optional[str] = None,
user_docker_base_url: Optional[str] = None,
unsafe_mode: bool = False,
**kwargs
):
"""
Initialize the code interpreter tool.
Args:
code (Optional[str]): Default code to execute
user_dockerfile_path (Optional[str]): Custom Dockerfile path
user_docker_base_url (Optional[str]): Custom Docker daemon URL
unsafe_mode (bool): Enable direct code execution
**kwargs: Additional arguments for base tool
"""
def _run(
self,
code: str,
libraries_used: List[str],
**kwargs: Any
) -> str:
"""
Execute Python code in Docker container or directly.
Args:
code (str): Python code to execute
libraries_used (List[str]): Required libraries
**kwargs: Additional arguments
Returns:
str: Execution output or error message
"""
```
## Best Practices
1. Security Considerations:
- Use Docker mode by default
- Validate input code
- Control library access
- Monitor execution time
2. Docker Configuration:
- Use custom Dockerfile when needed
- Handle container lifecycle
- Manage resource limits
- Clean up after execution
3. Library Management:
- Specify exact versions
- Use trusted packages
- Handle dependencies
- Verify installations
4. Error Handling:
- Catch execution errors
- Handle timeouts
- Manage Docker errors
- Provide clear messages
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import CodeInterpreterTool
# Initialize tool
code_tool = CodeInterpreterTool()
# Create agent
programmer = Agent(
role='Code Executor',
goal='Execute data analysis code',
backstory='Expert Python programmer specializing in data analysis.',
tools=[code_tool]
)
# Define task
analysis_task = Task(
description="""Analyze the dataset using pandas and
create a summary visualization with matplotlib.""",
agent=programmer
)
# The tool will use:
# {
# "code": """
# import pandas as pd
# import matplotlib.pyplot as plt
#
# # Load and analyze data
# df = pd.read_csv('data.csv')
# summary = df.describe()
#
# # Create visualization
# plt.figure(figsize=(10, 6))
# df['column'].hist()
# plt.savefig('output.png')
#
# print(summary)
# """,
# "libraries_used": "pandas,matplotlib"
# }
# Create crew
crew = Crew(
agents=[programmer],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Custom Docker Configuration
```python
# Use custom Dockerfile
tool = CodeInterpreterTool(
user_dockerfile_path="/path/to/Dockerfile"
)
# Use custom Docker daemon
tool = CodeInterpreterTool(
user_docker_base_url="tcp://remote-docker:2375"
)
```
### Direct Execution Mode
```python
# Enable unsafe mode (not recommended)
tool = CodeInterpreterTool(unsafe_mode=True)
# Execute code directly
result = tool.run(
code="print('Hello, World!')",
libraries_used=[]
)
```
### Error Handling Example
```python
try:
code_tool = CodeInterpreterTool()
result = code_tool.run(
code="""
import numpy as np
arr = np.array([1, 2, 3])
print(f"Array mean: {arr.mean()}")
""",
libraries_used=["numpy"]
)
print(result)
except Exception as e:
print(f"Error executing code: {str(e)}")
```
## Notes
- Inherits from BaseTool
- Docker-based isolation
- Dynamic library installation
- Secure code execution
- Custom Docker support
- Comprehensive error handling
- Resource management
- Container cleanup
- Library dependency handling
- Execution output capture

View File

@@ -0,0 +1,207 @@
---
title: CSVSearchTool
description: A tool for semantic search within CSV files using RAG capabilities
icon: table
---
## CSVSearchTool
The CSVSearchTool enables semantic search capabilities for CSV files using Retrieval-Augmented Generation (RAG). It can process CSV files either specified during initialization or at runtime, making it flexible for various use cases.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import CSVSearchTool
# Method 1: Initialize with specific CSV file
csv_tool = CSVSearchTool(csv="path/to/data.csv")
# Method 2: Initialize without CSV (specify at runtime)
flexible_csv_tool = CSVSearchTool()
# Create an agent with the tool
data_analyst = Agent(
role='Data Analyst',
goal='Search and analyze CSV data semantically',
backstory='Expert at analyzing and extracting insights from CSV data.',
tools=[csv_tool],
verbose=True
)
```
## Input Schema
### Fixed CSV Schema (when CSV path provided during initialization)
```python
class FixedCSVSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the CSV's content"
)
```
### Flexible CSV Schema (when CSV path provided at runtime)
```python
class CSVSearchToolSchema(FixedCSVSearchToolSchema):
csv: str = Field(
description="Mandatory csv path you want to search"
)
```
## Function Signature
```python
def __init__(
self,
csv: Optional[str] = None,
**kwargs
):
"""
Initialize the CSV search tool.
Args:
csv (Optional[str]): Path to CSV file (optional)
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on CSV content.
Args:
search_query (str): Query to search in the CSV
**kwargs: Additional arguments including csv path if not initialized
Returns:
str: Relevant content from the CSV matching the query
"""
```
## Best Practices
1. CSV File Handling:
- Ensure CSV files are properly formatted
- Use absolute paths for reliability
- Verify file permissions before processing
2. Search Optimization:
- Use specific, focused search queries
- Consider column names and data structure
- Test with sample queries first
3. Performance Considerations:
- Pre-initialize with CSV for repeated searches
- Handle large CSV files appropriately
- Monitor memory usage with big datasets
4. Error Handling:
- Verify CSV file existence
- Handle malformed CSV data
- Manage file access permissions
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import CSVSearchTool
# Initialize tool with specific CSV
csv_tool = CSVSearchTool(csv="/path/to/sales_data.csv")
# Create agent
analyst = Agent(
role='Data Analyst',
goal='Extract insights from sales data',
backstory='Expert at analyzing sales data and trends.',
tools=[csv_tool]
)
# Define task
analysis_task = Task(
description="""Find all sales records from the CSV
that relate to product returns in Q4 2023.""",
agent=analyst
)
# The tool will use:
# {
# "search_query": "product returns Q4 2023"
# }
# Create crew
crew = Crew(
agents=[analyst],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Dynamic CSV Selection
```python
# Initialize without CSV
flexible_tool = CSVSearchTool()
# Search different CSVs
result1 = flexible_tool.run(
search_query="revenue 2023",
csv="/path/to/finance.csv"
)
result2 = flexible_tool.run(
search_query="customer feedback",
csv="/path/to/surveys.csv"
)
```
### Multiple CSV Analysis
```python
# Create tools for different CSVs
sales_tool = CSVSearchTool(csv="/path/to/sales.csv")
inventory_tool = CSVSearchTool(csv="/path/to/inventory.csv")
# Create agent with multiple tools
analyst = Agent(
role='Business Analyst',
goal='Cross-reference sales and inventory data',
tools=[sales_tool, inventory_tool]
)
```
### Error Handling Example
```python
try:
csv_tool = CSVSearchTool(csv="/path/to/data.csv")
result = csv_tool.run(
search_query="important metrics"
)
print(result)
except Exception as e:
print(f"Error processing CSV: {str(e)}")
```
## Notes
- Inherits from RagTool for semantic search
- Supports dynamic CSV file specification
- Uses embedchain for data processing
- Maintains search context across queries
- Thread-safe operations
- Efficient semantic search capabilities
- Supports various CSV formats
- Handles large datasets effectively
- Preserves CSV structure in search
- Enables natural language queries

View File

@@ -0,0 +1,217 @@
---
title: Directory Read Tool
description: A tool for recursively listing directory contents
---
# Directory Read Tool
The Directory Read Tool provides functionality to recursively list all files within a directory. It supports both fixed and dynamic directory path modes, allowing you to specify the directory at initialization or runtime.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage
You can use the Directory Read Tool in two ways:
### 1. Fixed Directory Path
Initialize the tool with a specific directory path:
```python
from crewai import Agent
from crewai_tools import DirectoryReadTool
# Initialize with a fixed directory
tool = DirectoryReadTool(directory="/path/to/your/directory")
# Create an agent with the tool
agent = Agent(
role='File System Analyst',
goal='Analyze directory contents',
backstory='I help analyze and organize file systems',
tools=[tool]
)
# Use in a task
task = Task(
description="List all files in the project directory",
agent=agent
)
```
### 2. Dynamic Directory Path
Initialize the tool without a specific directory path to provide it at runtime:
```python
from crewai import Agent
from crewai_tools import DirectoryReadTool
# Initialize without a fixed directory
tool = DirectoryReadTool()
# Create an agent with the tool
agent = Agent(
role='File System Explorer',
goal='Explore different directories',
backstory='I analyze various directory structures',
tools=[tool]
)
# Use in a task with dynamic directory path
task = Task(
description="List all files in the specified directory",
agent=agent,
context={
"directory": "/path/to/explore"
}
)
```
## Input Schema
### Fixed Directory Mode
```python
class FixedDirectoryReadToolSchema(BaseModel):
pass # No additional parameters needed when directory is fixed
```
### Dynamic Directory Mode
```python
class DirectoryReadToolSchema(BaseModel):
directory: str # The path to the directory to list contents
```
## Function Signatures
```python
def __init__(self, directory: Optional[str] = None, **kwargs):
"""
Initialize the Directory Read Tool.
Args:
directory (Optional[str]): Path to the directory (optional)
**kwargs: Additional arguments passed to BaseTool
"""
def _run(
self,
**kwargs: Any,
) -> str:
"""
Execute the directory listing.
Args:
**kwargs: Arguments including 'directory' for dynamic mode
Returns:
str: A formatted string containing all file paths in the directory
"""
```
## Best Practices
1. **Path Handling**:
- Use absolute paths to avoid path resolution issues
- Handle trailing slashes appropriately
- Verify directory existence before listing
2. **Performance Considerations**:
- Be mindful of directory size when listing large directories
- Consider implementing pagination for large directories
- Handle symlinks appropriately
3. **Error Handling**:
- Handle directory not found errors gracefully
- Manage permission issues appropriately
- Validate input parameters before processing
## Example Integration
Here's a complete example showing how to integrate the Directory Read Tool with CrewAI:
```python
from crewai import Agent, Task, Crew
from crewai_tools import DirectoryReadTool
# Initialize the tool
dir_tool = DirectoryReadTool()
# Create an agent with the tool
file_analyst = Agent(
role='File System Analyst',
goal='Analyze and report on directory structures',
backstory='I am an expert at analyzing file system organization',
tools=[dir_tool]
)
# Create tasks
analysis_task = Task(
description="""
Analyze the project directory structure:
1. List all files recursively
2. Identify key file types
3. Report on directory organization
Provide a comprehensive analysis of the findings.
""",
agent=file_analyst,
context={
"directory": "/path/to/project"
}
)
# Create and run the crew
crew = Crew(
agents=[file_analyst],
tasks=[analysis_task]
)
result = crew.kickoff()
```
## Error Handling
The tool handles various error scenarios:
1. **Directory Not Found**:
```python
try:
tool = DirectoryReadTool(directory="/nonexistent/path")
except FileNotFoundError:
print("Directory not found. Please verify the path.")
```
2. **Permission Issues**:
```python
try:
tool = DirectoryReadTool(directory="/restricted/path")
except PermissionError:
print("Insufficient permissions to access the directory.")
```
3. **Invalid Path**:
```python
try:
result = tool._run(directory="invalid/path")
except ValueError:
print("Invalid directory path provided.")
```
## Output Format
The tool returns a formatted string containing all file paths in the directory:
```
File paths:
- /path/to/directory/file1.txt
- /path/to/directory/subdirectory/file2.txt
- /path/to/directory/subdirectory/file3.py
```
Each file path is listed on a new line with a hyphen prefix, making it easy to parse and read the output.

View File

@@ -0,0 +1,214 @@
---
title: DirectorySearchTool
description: A tool for semantic search within directory contents using RAG capabilities
icon: folder-search
---
## DirectorySearchTool
The DirectorySearchTool enables semantic search capabilities for directory contents using Retrieval-Augmented Generation (RAG). It processes files recursively within a directory and allows searching through their contents using natural language queries.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import DirectorySearchTool
# Method 1: Initialize with specific directory
dir_tool = DirectorySearchTool(directory="/path/to/documents")
# Method 2: Initialize without directory (specify at runtime)
flexible_dir_tool = DirectorySearchTool()
# Create an agent with the tool
researcher = Agent(
role='Directory Researcher',
goal='Search and analyze directory contents',
backstory='Expert at finding relevant information in document collections.',
tools=[dir_tool],
verbose=True
)
```
## Input Schema
### Fixed Directory Schema (when path provided during initialization)
```python
class FixedDirectorySearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the directory's content"
)
```
### Flexible Directory Schema (when path provided at runtime)
```python
class DirectorySearchToolSchema(FixedDirectorySearchToolSchema):
directory: str = Field(
description="Mandatory directory you want to search"
)
```
## Function Signature
```python
def __init__(
self,
directory: Optional[str] = None,
**kwargs
):
"""
Initialize the directory search tool.
Args:
directory (Optional[str]): Path to directory (optional)
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on directory contents.
Args:
search_query (str): Query to search in the directory
**kwargs: Additional arguments including directory if not initialized
Returns:
str: Relevant content from the directory matching the query
"""
```
## Best Practices
1. Directory Management:
- Use absolute paths
- Verify directory existence
- Handle permissions properly
2. Search Optimization:
- Use specific queries
- Consider file types
- Test with sample queries
3. Performance Considerations:
- Pre-initialize for repeated searches
- Handle large directories
- Monitor processing time
4. Error Handling:
- Verify directory access
- Handle missing files
- Manage permissions
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import DirectorySearchTool
# Initialize tool with specific directory
dir_tool = DirectorySearchTool(
directory="/path/to/documents"
)
# Create agent
researcher = Agent(
role='Directory Researcher',
goal='Extract insights from document collections',
backstory='Expert at analyzing document collections.',
tools=[dir_tool]
)
# Define task
research_task = Task(
description="""Find all mentions of machine learning
applications from the directory contents.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "machine learning applications"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Dynamic Directory Selection
```python
# Initialize without directory path
flexible_tool = DirectorySearchTool()
# Search different directories
docs_results = flexible_tool.run(
search_query="technical specifications",
directory="/path/to/docs"
)
reports_results = flexible_tool.run(
search_query="financial metrics",
directory="/path/to/reports"
)
```
### Multiple Directory Analysis
```python
# Create tools for different directories
docs_tool = DirectorySearchTool(
directory="/path/to/docs"
)
reports_tool = DirectorySearchTool(
directory="/path/to/reports"
)
# Create agent with multiple tools
analyst = Agent(
role='Content Analyst',
goal='Cross-reference multiple document collections',
tools=[docs_tool, reports_tool]
)
```
### Error Handling Example
```python
try:
dir_tool = DirectorySearchTool()
results = dir_tool.run(
search_query="key concepts",
directory="/path/to/documents"
)
print(results)
except Exception as e:
print(f"Error processing directory: {str(e)}")
```
## Notes
- Inherits from RagTool
- Uses DirectoryLoader
- Supports recursive search
- Dynamic directory specification
- Efficient content retrieval
- Thread-safe operations
- Maintains search context
- Processes multiple file types
- Handles nested directories
- Memory-efficient processing

View File

@@ -0,0 +1,224 @@
---
title: DOCXSearchTool
description: A tool for semantic search within DOCX documents using RAG capabilities
icon: file-text
---
## DOCXSearchTool
The DOCXSearchTool enables semantic search capabilities for Microsoft Word (DOCX) documents using Retrieval-Augmented Generation (RAG). It supports both fixed and dynamic document selection modes.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import DOCXSearchTool
# Method 1: Fixed document (specified at initialization)
fixed_tool = DOCXSearchTool(
docx="path/to/document.docx"
)
# Method 2: Dynamic document (specified at runtime)
dynamic_tool = DOCXSearchTool()
# Create an agent with the tool
researcher = Agent(
role='Document Researcher',
goal='Search and analyze document contents',
backstory='Expert at finding relevant information in documents.',
tools=[fixed_tool], # or [dynamic_tool]
verbose=True
)
```
## Input Schema
### Fixed Document Mode
```python
class FixedDOCXSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the DOCX's content"
)
```
### Dynamic Document Mode
```python
class DOCXSearchToolSchema(BaseModel):
docx: str = Field(
description="Mandatory docx path you want to search"
)
search_query: str = Field(
description="Mandatory search query you want to use to search the DOCX's content"
)
```
## Function Signature
```python
def __init__(
self,
docx: Optional[str] = None,
**kwargs
):
"""
Initialize the DOCX search tool.
Args:
docx (Optional[str]): Path to DOCX file (optional for dynamic mode)
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
docx: Optional[str] = None,
**kwargs: Any
) -> str:
"""
Execute semantic search on document contents.
Args:
search_query (str): Query to search in the document
docx (Optional[str]): Document path (required for dynamic mode)
**kwargs: Additional arguments
Returns:
str: Relevant content from the document matching the query
"""
```
## Best Practices
1. Document Handling:
- Use absolute file paths
- Verify file existence
- Handle large documents
- Monitor memory usage
2. Query Optimization:
- Structure queries clearly
- Consider document size
- Handle formatting
- Monitor performance
3. Error Handling:
- Check file access
- Validate file format
- Handle corrupted files
- Log issues
4. Mode Selection:
- Choose fixed mode for static documents
- Use dynamic mode for runtime selection
- Consider memory implications
- Manage document lifecycle
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import DOCXSearchTool
# Initialize tool
docx_tool = DOCXSearchTool(
docx="reports/annual_report_2023.docx"
)
# Create agent
researcher = Agent(
role='Document Analyst',
goal='Extract insights from annual report',
backstory='Expert at analyzing business documents.',
tools=[docx_tool]
)
# Define task
analysis_task = Task(
description="""Find all mentions of revenue growth
and market expansion.""",
agent=researcher
)
# Create crew
crew = Crew(
agents=[researcher],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Multiple Document Analysis
```python
# Create tools for different documents
report_tool = DOCXSearchTool(
docx="reports/annual_report.docx"
)
policy_tool = DOCXSearchTool(
docx="policies/compliance.docx"
)
# Create agent with multiple tools
analyst = Agent(
role='Document Analyst',
goal='Cross-reference reports and policies',
tools=[report_tool, policy_tool]
)
```
### Dynamic Document Loading
```python
# Initialize dynamic tool
dynamic_tool = DOCXSearchTool()
# Use with different documents
result1 = dynamic_tool.run(
docx="document1.docx",
search_query="project timeline"
)
result2 = dynamic_tool.run(
docx="document2.docx",
search_query="budget allocation"
)
```
### Error Handling Example
```python
try:
docx_tool = DOCXSearchTool(
docx="reports/quarterly_report.docx"
)
results = docx_tool.run(
search_query="Q3 performance metrics"
)
print(results)
except FileNotFoundError as e:
print(f"Document not found: {str(e)}")
except Exception as e:
print(f"Error processing document: {str(e)}")
```
## Notes
- Inherits from RagTool
- Supports fixed/dynamic modes
- Document path validation
- Memory management
- Performance optimization
- Error handling
- Search capabilities
- Content extraction
- Format handling
- Security features

View File

@@ -0,0 +1,193 @@
---
title: FileReadTool
description: A tool for reading file contents with flexible path specification
icon: file-text
---
## FileReadTool
The FileReadTool provides functionality to read file contents with support for both fixed and dynamic file path specification. It includes comprehensive error handling for common file operations and maintains clear descriptions of its configured state.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import FileReadTool
# Method 1: Initialize with specific file
reader = FileReadTool(file_path="/path/to/data.txt")
# Method 2: Initialize without file (specify at runtime)
flexible_reader = FileReadTool()
# Create an agent with the tool
file_processor = Agent(
role='File Processor',
goal='Read and process file contents',
backstory='Expert at handling file operations and content processing.',
tools=[reader],
verbose=True
)
```
## Input Schema
```python
class FileReadToolSchema(BaseModel):
file_path: str = Field(
description="Mandatory file full path to read the file"
)
```
## Function Signature
```python
def __init__(
self,
file_path: Optional[str] = None,
**kwargs: Any
) -> None:
"""
Initialize the file read tool.
Args:
file_path (Optional[str]): Path to file to read (optional)
**kwargs: Additional arguments passed to BaseTool
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Read and return file contents.
Args:
file_path (str, optional): Override default file path
**kwargs: Additional arguments
Returns:
str: File contents or error message
"""
```
## Best Practices
1. File Path Management:
- Use absolute paths for reliability
- Verify file existence before operations
- Handle path resolution properly
2. Error Handling:
- Check for file existence
- Handle permission issues
- Manage encoding errors
- Process file access failures
3. Performance Considerations:
- Close files after reading
- Handle large files appropriately
- Consider memory constraints
4. Security Practices:
- Validate file paths
- Check file permissions
- Avoid path traversal issues
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import FileReadTool
# Initialize tool with specific file
reader = FileReadTool(file_path="/path/to/config.txt")
# Create agent
processor = Agent(
role='File Processor',
goal='Process configuration files',
backstory='Expert at reading and analyzing configuration files.',
tools=[reader]
)
# Define task
read_task = Task(
description="""Read and analyze the contents of
the configuration file.""",
agent=processor
)
# The tool will use the default file path
# Create crew
crew = Crew(
agents=[processor],
tasks=[read_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Dynamic File Selection
```python
# Initialize without file path
flexible_reader = FileReadTool()
# Read different files
config_content = flexible_reader.run(
file_path="/path/to/config.txt"
)
log_content = flexible_reader.run(
file_path="/path/to/logs.txt"
)
```
### Multiple File Processing
```python
# Create tools for different files
config_reader = FileReadTool(file_path="/path/to/config.txt")
log_reader = FileReadTool(file_path="/path/to/logs.txt")
# Create agent with multiple tools
processor = Agent(
role='File Analyst',
goal='Analyze multiple file types',
tools=[config_reader, log_reader]
)
```
### Error Handling Example
```python
try:
reader = FileReadTool()
content = reader.run(
file_path="/path/to/file.txt"
)
print(content)
except Exception as e:
print(f"Error reading file: {str(e)}")
```
## Notes
- Inherits from BaseTool
- Supports fixed or dynamic file paths
- Comprehensive error handling
- Thread-safe operations
- Clear error messages
- Flexible path specification
- Maintains tool description
- Handles common file errors
- Supports various file types
- Memory-efficient operations

View File

@@ -0,0 +1,141 @@
---
title: FileWriterTool
description: A tool for writing content to files with support for various file formats.
icon: file-pen
---
## FileWriterTool
The FileWriterTool provides agents with the capability to write content to files, supporting various file formats and ensuring proper file handling.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import FileWriterTool
# Initialize the tool
file_writer = FileWriterTool()
# Create an agent with the tool
writer_agent = Agent(
role='Content Writer',
goal='Write and save content to files',
backstory='Expert at creating and managing file content.',
tools=[file_writer],
verbose=True
)
# Use in a task
task = Task(
description='Write a report and save it to report.txt',
agent=writer_agent
)
```
## Tool Attributes
| Attribute | Type | Description |
| :-------- | :--- | :---------- |
| name | str | "File Writer Tool" |
| description | str | "A tool that writes content to a file." |
## Input Schema
```python
class FileWriterToolInput(BaseModel):
filename: str # Name of the file to write
directory: str = "./" # Optional directory path, defaults to current directory
overwrite: str = "False" # Whether to overwrite existing file ("True"/"False")
content: str # Content to write to the file
```
## Function Signature
```python
def _run(self, **kwargs: Any) -> str:
"""
Write content to a file with specified parameters.
Args:
filename (str): Name of the file to write
content (str): Content to write to the file
directory (str, optional): Directory path. Defaults to "./".
overwrite (str, optional): Whether to overwrite existing file. Defaults to "False".
Returns:
str: Success message with filepath or error message
"""
```
## Error Handling
The tool includes error handling for common file operations:
- FileExistsError: When file exists and overwrite is not allowed
- KeyError: When required parameters are missing
- Directory Creation: Automatically creates directories if they don't exist
- General Exceptions: Catches and reports any other file operation errors
## Best Practices
1. Always provide absolute file paths
2. Ensure proper file permissions
3. Handle potential errors in your agent prompts
4. Verify file contents after writing
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import FileWriterTool
# Initialize tool
file_writer = FileWriterTool()
# Create agent
writer = Agent(
role='Technical Writer',
goal='Create and save technical documentation',
backstory='Expert technical writer with experience in documentation.',
tools=[file_writer]
)
# Define task
writing_task = Task(
description="""Write a technical guide about Python best practices and save it
to the docs directory. The file should be named 'python_guide.md'.
Include sections on code style, documentation, and testing.
If a file already exists, overwrite it.""",
agent=writer
)
# The agent can use the tool with these parameters:
# {
# "filename": "python_guide.md",
# "directory": "docs",
# "overwrite": "True",
# "content": "# Python Best Practices\n\n## Code Style\n..."
# }
# Create crew
crew = Crew(
agents=[writer],
tasks=[writing_task]
)
# Execute
result = crew.kickoff()
```
## Notes
- The tool automatically creates directories in the file path if they don't exist
- Supports various file formats (txt, md, json, etc.)
- Returns descriptive error messages for better debugging
- Thread-safe file operations

View File

@@ -0,0 +1,181 @@
---
title: FirecrawlCrawlWebsiteTool
description: A web crawling tool powered by Firecrawl API for comprehensive website content extraction
icon: spider-web
---
## FirecrawlCrawlWebsiteTool
The FirecrawlCrawlWebsiteTool provides website crawling capabilities using the Firecrawl API. It allows for customizable crawling with options for polling intervals, idempotency, and URL parameters.
## Installation
```bash
pip install 'crewai[tools]'
pip install firecrawl-py # Required dependency
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import FirecrawlCrawlWebsiteTool
# Method 1: Using environment variable
# export FIRECRAWL_API_KEY='your-api-key'
crawler = FirecrawlCrawlWebsiteTool()
# Method 2: Providing API key directly
crawler = FirecrawlCrawlWebsiteTool(
api_key="your-firecrawl-api-key"
)
# Method 3: With custom configuration
crawler = FirecrawlCrawlWebsiteTool(
api_key="your-firecrawl-api-key",
url="https://example.com", # Base URL
poll_interval=5, # Custom polling interval
idempotency_key="unique-key"
)
# Create an agent with the tool
researcher = Agent(
role='Web Crawler',
goal='Extract and analyze website content',
backstory='Expert at crawling and analyzing web content.',
tools=[crawler],
verbose=True
)
```
## Input Schema
```python
class FirecrawlCrawlWebsiteToolSchema(BaseModel):
url: str = Field(description="Website URL")
```
## Function Signature
```python
def __init__(
self,
api_key: Optional[str] = None,
url: Optional[str] = None,
params: Optional[Dict[str, Any]] = None,
poll_interval: Optional[int] = 2,
idempotency_key: Optional[str] = None,
**kwargs
):
"""
Initialize the website crawling tool.
Args:
api_key (Optional[str]): Firecrawl API key. If not provided, checks FIRECRAWL_API_KEY env var
url (Optional[str]): Base URL to crawl. Can be overridden in _run
params (Optional[Dict[str, Any]]): Additional parameters for FirecrawlApp
poll_interval (Optional[int]): Poll interval for FirecrawlApp
idempotency_key (Optional[str]): Idempotency key for FirecrawlApp
**kwargs: Additional arguments for tool creation
"""
def _run(self, url: str) -> Any:
"""
Crawl a website using Firecrawl.
Args:
url (str): Website URL to crawl (overrides constructor URL if provided)
Returns:
Any: Crawled website content from Firecrawl API
"""
```
## Best Practices
1. Set up API authentication:
- Use environment variable: `export FIRECRAWL_API_KEY='your-api-key'`
- Or provide directly in constructor
2. Configure crawling parameters:
- Set appropriate poll intervals
- Use idempotency keys for retry safety
- Customize URL parameters as needed
3. Handle rate limits and quotas
4. Consider website robots.txt policies
5. Handle potential crawling errors in agent prompts
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import FirecrawlCrawlWebsiteTool
# Initialize crawler with configuration
crawler = FirecrawlCrawlWebsiteTool(
api_key="your-firecrawl-api-key",
poll_interval=5,
params={
"max_depth": 3,
"follow_links": True
}
)
# Create agent
web_analyst = Agent(
role='Web Content Analyst',
goal='Extract and analyze website content comprehensively',
backstory='Expert at web crawling and content analysis.',
tools=[crawler]
)
# Define task
crawl_task = Task(
description="""Crawl the documentation website at docs.example.com
and extract all API-related content.""",
agent=web_analyst
)
# The agent will use:
# {
# "url": "https://docs.example.com"
# }
# Create crew
crew = Crew(
agents=[web_analyst],
tasks=[crawl_task]
)
# Execute
result = crew.kickoff()
```
## Configuration Options
### URL Parameters
```python
params = {
"max_depth": 3, # Maximum crawl depth
"follow_links": True, # Follow internal links
"exclude_patterns": [], # URL patterns to exclude
"include_patterns": [] # URL patterns to include
}
```
### Polling Configuration
```python
crawler = FirecrawlCrawlWebsiteTool(
poll_interval=5, # Poll every 5 seconds
idempotency_key="unique-key-123" # For retry safety
)
```
## Notes
- Requires valid Firecrawl API key
- Supports both environment variable and direct API key configuration
- Configurable polling intervals for crawl status
- Idempotency support for safe retries
- Thread-safe operations
- Customizable crawling parameters
- Respects robots.txt by default

View File

@@ -0,0 +1,154 @@
---
title: FirecrawlSearchTool
description: A web search tool powered by Firecrawl API for comprehensive web search capabilities
icon: magnifying-glass-chart
---
## FirecrawlSearchTool
The FirecrawlSearchTool provides web search capabilities using the Firecrawl API. It allows for customizable search queries with options for result formatting and search parameters.
## Installation
```bash
pip install 'crewai[tools]'
pip install firecrawl-py # Required dependency
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import FirecrawlSearchTool
# Initialize the tool with your API key
search_tool = FirecrawlSearchTool(api_key="your-firecrawl-api-key")
# Create an agent with the tool
researcher = Agent(
role='Web Researcher',
goal='Find relevant information across the web',
backstory='Expert at web research and information gathering.',
tools=[search_tool],
verbose=True
)
```
## Input Schema
```python
class FirecrawlSearchToolSchema(BaseModel):
query: str = Field(description="Search query")
page_options: Optional[Dict[str, Any]] = Field(
default=None,
description="Options for result formatting"
)
search_options: Optional[Dict[str, Any]] = Field(
default=None,
description="Options for searching"
)
```
## Function Signature
```python
def __init__(self, api_key: Optional[str] = None, **kwargs):
"""
Initialize the Firecrawl search tool.
Args:
api_key (Optional[str]): Firecrawl API key
**kwargs: Additional arguments for tool creation
"""
def _run(
self,
query: str,
page_options: Optional[Dict[str, Any]] = None,
result_options: Optional[Dict[str, Any]] = None,
) -> Any:
"""
Perform a web search using Firecrawl.
Args:
query (str): Search query string
page_options (Optional[Dict[str, Any]]): Options for result formatting
result_options (Optional[Dict[str, Any]]): Options for search results
Returns:
Any: Search results from Firecrawl API
"""
```
## Best Practices
1. Always provide a valid API key
2. Use specific, focused search queries
3. Customize page and result options for better results
4. Handle potential API errors in agent prompts
5. Consider rate limits and usage quotas
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import FirecrawlSearchTool
# Initialize tool with API key
search_tool = FirecrawlSearchTool(api_key="your-firecrawl-api-key")
# Create agent
researcher = Agent(
role='Market Researcher',
goal='Research market trends and competitor analysis',
backstory='Expert market analyst with deep research skills.',
tools=[search_tool]
)
# Define task
research_task = Task(
description="""Research the latest developments in electric vehicles,
focusing on market leaders and emerging technologies. Format the results
in a structured way.""",
agent=researcher
)
# The agent will use:
# {
# "query": "electric vehicle market leaders emerging technologies",
# "page_options": {
# "format": "structured",
# "maxLength": 1000
# },
# "result_options": {
# "limit": 5,
# "sortBy": "relevance"
# }
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Error Handling
The tool includes error handling for:
- Missing API key
- Missing firecrawl-py package
- API request failures
- Invalid options parameters
## Notes
- Requires valid Firecrawl API key
- Supports customizable search parameters
- Provides structured web search results
- Thread-safe operations
- Efficient for large-scale web searches
- Handles rate limiting automatically

View File

@@ -0,0 +1,233 @@
---
title: GithubSearchTool
description: A tool for semantic search within GitHub repositories using RAG capabilities
icon: github
---
## GithubSearchTool
The GithubSearchTool enables semantic search capabilities for GitHub repositories using Retrieval-Augmented Generation (RAG). It processes various content types including code, repository information, pull requests, and issues, allowing natural language queries across repository content.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import GithubSearchTool
# Method 1: Initialize with specific repository
github_tool = GithubSearchTool(
github_repo="owner/repo",
gh_token="your_github_token",
content_types=["code", "pr", "issue"]
)
# Method 2: Initialize without repository (specify at runtime)
flexible_github_tool = GithubSearchTool(
gh_token="your_github_token",
content_types=["code", "repo"]
)
# Create an agent with the tool
researcher = Agent(
role='GitHub Researcher',
goal='Search and analyze repository contents',
backstory='Expert at finding relevant information in GitHub repositories.',
tools=[github_tool],
verbose=True
)
```
## Input Schema
### Fixed Repository Schema (when repo provided during initialization)
```python
class FixedGithubSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the github repo's content"
)
```
### Flexible Repository Schema (when repo provided at runtime)
```python
class GithubSearchToolSchema(FixedGithubSearchToolSchema):
github_repo: str = Field(
description="Mandatory github you want to search"
)
content_types: List[str] = Field(
description="Mandatory content types you want to be included search, options: [code, repo, pr, issue]"
)
```
## Function Signature
```python
def __init__(
self,
github_repo: Optional[str] = None,
gh_token: str,
content_types: List[str],
**kwargs
):
"""
Initialize the GitHub search tool.
Args:
github_repo (Optional[str]): Repository to search (optional)
gh_token (str): GitHub authentication token
content_types (List[str]): Content types to search
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on repository contents.
Args:
search_query (str): Query to search in the repository
**kwargs: Additional arguments including github_repo and content_types if not initialized
Returns:
str: Relevant content from the repository matching the query
"""
```
## Best Practices
1. Authentication:
- Secure token management
- Use environment variables
- Handle token expiration
2. Search Optimization:
- Target specific content types
- Use focused queries
- Consider rate limits
3. Performance Considerations:
- Pre-initialize for repeated searches
- Handle large repositories
- Monitor API usage
4. Error Handling:
- Verify repository access
- Handle API limits
- Manage authentication errors
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import GithubSearchTool
# Initialize tool with specific repository
github_tool = GithubSearchTool(
github_repo="owner/repo",
gh_token="your_github_token",
content_types=["code", "pr", "issue"]
)
# Create agent
researcher = Agent(
role='GitHub Researcher',
goal='Extract insights from repository content',
backstory='Expert at analyzing GitHub repositories.',
tools=[github_tool]
)
# Define task
research_task = Task(
description="""Find all implementations of
machine learning algorithms in the codebase.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "machine learning implementation"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Dynamic Repository Selection
```python
# Initialize without repository
flexible_tool = GithubSearchTool(
gh_token="your_github_token",
content_types=["code", "repo"]
)
# Search different repositories
backend_results = flexible_tool.run(
search_query="authentication implementation",
github_repo="owner/backend-repo"
)
frontend_results = flexible_tool.run(
search_query="component architecture",
github_repo="owner/frontend-repo"
)
```
### Multiple Content Type Analysis
```python
# Create tool with multiple content types
multi_tool = GithubSearchTool(
github_repo="owner/repo",
gh_token="your_github_token",
content_types=["code", "pr", "issue", "repo"]
)
# Search across all content types
results = multi_tool.run(
search_query="feature implementation status"
)
```
### Error Handling Example
```python
try:
github_tool = GithubSearchTool(
gh_token="your_github_token",
content_types=["code"]
)
results = github_tool.run(
search_query="api endpoints",
github_repo="owner/repo"
)
print(results)
except Exception as e:
print(f"Error searching repository: {str(e)}")
```
## Notes
- Inherits from RagTool
- Uses GithubLoader
- Requires authentication
- Supports multiple content types
- Dynamic repository specification
- Efficient content retrieval
- Thread-safe operations
- Maintains search context
- Handles API rate limits
- Memory-efficient processing

View File

@@ -0,0 +1,220 @@
---
title: JinaScrapeWebsiteTool
description: A tool for scraping website content using Jina.ai's reader service with markdown output
icon: globe
---
## JinaScrapeWebsiteTool
The JinaScrapeWebsiteTool provides website content scraping capabilities using Jina.ai's reader service. It converts web content into clean markdown format and supports both fixed and dynamic URL modes with optional authentication.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import JinaScrapeWebsiteTool
# Method 1: Fixed URL (specified at initialization)
fixed_tool = JinaScrapeWebsiteTool(
website_url="https://example.com",
api_key="your-jina-api-key" # Optional
)
# Method 2: Dynamic URL (specified at runtime)
dynamic_tool = JinaScrapeWebsiteTool(
api_key="your-jina-api-key" # Optional
)
# Create an agent with the tool
researcher = Agent(
role='Web Content Researcher',
goal='Extract and analyze website content',
backstory='Expert at gathering and processing web information.',
tools=[fixed_tool], # or [dynamic_tool]
verbose=True
)
```
## Input Schema
```python
class JinaScrapeWebsiteToolInput(BaseModel):
website_url: str = Field(
description="Mandatory website url to read the file"
)
```
## Function Signature
```python
def __init__(
self,
website_url: Optional[str] = None,
api_key: Optional[str] = None,
custom_headers: Optional[dict] = None,
**kwargs
):
"""
Initialize the website scraping tool.
Args:
website_url (Optional[str]): URL to scrape (optional for dynamic mode)
api_key (Optional[str]): Jina.ai API key for authentication
custom_headers (Optional[dict]): Custom HTTP headers
**kwargs: Additional arguments for base tool
"""
def _run(
self,
website_url: Optional[str] = None
) -> str:
"""
Execute website scraping.
Args:
website_url (Optional[str]): URL to scrape (required for dynamic mode)
Returns:
str: Markdown-formatted website content
"""
```
## Best Practices
1. URL Handling:
- Use complete URLs
- Validate URL format
- Handle redirects
- Monitor timeouts
2. Authentication:
- Secure API key storage
- Use environment variables
- Manage headers properly
- Handle auth errors
3. Content Processing:
- Handle large pages
- Process markdown output
- Manage encoding
- Handle errors
4. Mode Selection:
- Choose fixed mode for static sites
- Use dynamic mode for variable URLs
- Consider caching
- Manage timeouts
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import JinaScrapeWebsiteTool
import os
# Initialize tool with API key
scraper_tool = JinaScrapeWebsiteTool(
api_key=os.getenv('JINA_API_KEY'),
custom_headers={
'User-Agent': 'CrewAI Bot 1.0'
}
)
# Create agent
researcher = Agent(
role='Web Content Analyst',
goal='Extract and analyze website content',
backstory='Expert at processing web information.',
tools=[scraper_tool]
)
# Define task
analysis_task = Task(
description="""Analyze the content of
https://example.com/blog for key insights.""",
agent=researcher
)
# Create crew
crew = Crew(
agents=[researcher],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Multiple Site Analysis
```python
# Initialize tool
scraper = JinaScrapeWebsiteTool(
api_key=os.getenv('JINA_API_KEY')
)
# Analyze multiple sites
results = []
sites = [
"https://site1.com",
"https://site2.com",
"https://site3.com"
]
for site in sites:
content = scraper.run(
website_url=site
)
results.append(content)
```
### Custom Headers Configuration
```python
# Initialize with custom headers
tool = JinaScrapeWebsiteTool(
custom_headers={
'User-Agent': 'Custom Bot 1.0',
'Accept-Language': 'en-US,en;q=0.9',
'Accept': 'text/html,application/xhtml+xml'
}
)
# Use the tool
content = tool.run(
website_url="https://example.com"
)
```
### Error Handling Example
```python
try:
scraper = JinaScrapeWebsiteTool()
content = scraper.run(
website_url="https://example.com"
)
print(content)
except requests.exceptions.RequestException as e:
print(f"Error accessing website: {str(e)}")
except Exception as e:
print(f"Error processing content: {str(e)}")
```
## Notes
- Uses Jina.ai reader service
- Markdown output format
- API key authentication
- Custom headers support
- Error handling
- Timeout management
- Content processing
- URL validation
- Redirect handling
- Response formatting

View File

@@ -0,0 +1,224 @@
---
title: JSONSearchTool
description: A tool for semantic search within JSON files using RAG capabilities
icon: braces
---
## JSONSearchTool
The JSONSearchTool enables semantic search capabilities for JSON files using Retrieval-Augmented Generation (RAG). It supports both fixed and dynamic file path modes, allowing flexible usage patterns.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import JSONSearchTool
# Method 1: Fixed path (specified at initialization)
fixed_tool = JSONSearchTool(
json_path="path/to/data.json"
)
# Method 2: Dynamic path (specified at runtime)
dynamic_tool = JSONSearchTool()
# Create an agent with the tool
researcher = Agent(
role='JSON Data Researcher',
goal='Search and analyze JSON data',
backstory='Expert at finding relevant information in JSON files.',
tools=[fixed_tool], # or [dynamic_tool]
verbose=True
)
```
## Input Schema
### Fixed Path Mode
```python
class FixedJSONSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the JSON's content"
)
```
### Dynamic Path Mode
```python
class JSONSearchToolSchema(BaseModel):
json_path: str = Field(
description="Mandatory json path you want to search"
)
search_query: str = Field(
description="Mandatory search query you want to use to search the JSON's content"
)
```
## Function Signature
```python
def __init__(
self,
json_path: Optional[str] = None,
**kwargs
):
"""
Initialize the JSON search tool.
Args:
json_path (Optional[str]): Path to JSON file (optional for dynamic mode)
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on JSON contents.
Args:
search_query (str): Query to search in the JSON
**kwargs: Additional arguments
Returns:
str: Relevant content from the JSON matching the query
"""
```
## Best Practices
1. File Handling:
- Use absolute file paths
- Verify file existence
- Handle large JSON files
- Monitor memory usage
2. Query Optimization:
- Structure queries clearly
- Consider JSON structure
- Handle nested data
- Monitor performance
3. Error Handling:
- Check file access
- Validate JSON format
- Handle malformed JSON
- Log issues
4. Mode Selection:
- Choose fixed mode for static files
- Use dynamic mode for runtime selection
- Consider caching
- Manage file lifecycle
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import JSONSearchTool
# Initialize tool
json_tool = JSONSearchTool(
json_path="data/config.json"
)
# Create agent
researcher = Agent(
role='JSON Data Analyst',
goal='Extract insights from JSON configuration',
backstory='Expert at analyzing JSON data structures.',
tools=[json_tool]
)
# Define task
analysis_task = Task(
description="""Find all configuration settings
related to security.""",
agent=researcher
)
# Create crew
crew = Crew(
agents=[researcher],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Multiple File Analysis
```python
# Create tools for different JSON files
config_tool = JSONSearchTool(
json_path="config/settings.json"
)
data_tool = JSONSearchTool(
json_path="data/records.json"
)
# Create agent with multiple tools
analyst = Agent(
role='JSON Data Analyst',
goal='Cross-reference configuration and data',
tools=[config_tool, data_tool]
)
```
### Dynamic File Loading
```python
# Initialize dynamic tool
dynamic_tool = JSONSearchTool()
# Use with different JSON files
result1 = dynamic_tool.run(
json_path="file1.json",
search_query="security settings"
)
result2 = dynamic_tool.run(
json_path="file2.json",
search_query="user preferences"
)
```
### Error Handling Example
```python
try:
json_tool = JSONSearchTool(
json_path="config/settings.json"
)
results = json_tool.run(
search_query="encryption settings"
)
print(results)
except FileNotFoundError as e:
print(f"JSON file not found: {str(e)}")
except ValueError as e:
print(f"Invalid JSON format: {str(e)}")
except Exception as e:
print(f"Error processing JSON: {str(e)}")
```
## Notes
- Inherits from RagTool
- Supports fixed/dynamic modes
- JSON path validation
- Memory management
- Performance optimization
- Error handling
- Search capabilities
- Content extraction
- Format validation
- Security features

View File

@@ -0,0 +1,184 @@
---
title: LinkupSearchTool
description: A search tool powered by Linkup API for retrieving contextual information
icon: search
---
## LinkupSearchTool
The LinkupSearchTool provides search capabilities using the Linkup API. It allows for customizable search depth and output formatting, returning structured results with contextual information.
## Installation
```bash
pip install 'crewai[tools]'
pip install linkup # Required dependency
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import LinkupSearchTool
# Initialize the tool with your API key
search_tool = LinkupSearchTool(api_key="your-linkup-api-key")
# Create an agent with the tool
researcher = Agent(
role='Information Researcher',
goal='Find relevant contextual information',
backstory='Expert at retrieving and analyzing contextual data.',
tools=[search_tool],
verbose=True
)
```
## Function Signature
```python
def __init__(self, api_key: str):
"""
Initialize the Linkup search tool.
Args:
api_key (str): Linkup API key for authentication
"""
def _run(
self,
query: str,
depth: str = "standard",
output_type: str = "searchResults"
) -> dict:
"""
Perform a search using the Linkup API.
Args:
query (str): The search query
depth (str): Search depth ("standard" by default)
output_type (str): Desired result type ("searchResults" by default)
Returns:
dict: {
"success": bool,
"results": List[Dict] | None,
"error": str | None
}
On success, results contains list of:
{
"name": str,
"url": str,
"content": str
}
"""
```
## Best Practices
1. Always provide a valid API key
2. Use specific, focused search queries
3. Choose appropriate search depth based on needs
4. Handle potential API errors in agent prompts
5. Process structured results effectively
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import LinkupSearchTool
# Initialize tool with API key
search_tool = LinkupSearchTool(api_key="your-linkup-api-key")
# Create agent
researcher = Agent(
role='Context Researcher',
goal='Find detailed contextual information about topics',
backstory='Expert at discovering and analyzing contextual data.',
tools=[search_tool]
)
# Define task
research_task = Task(
description="""Research the latest developments in quantum computing,
focusing on recent breakthroughs and applications. Use standard depth
for comprehensive results.""",
agent=researcher
)
# The tool will use:
# query: "quantum computing recent breakthroughs applications"
# depth: "standard"
# output_type: "searchResults"
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Search Depth Options
```python
# Quick surface-level search
results = search_tool._run(
query="quantum computing",
depth="basic"
)
# Standard comprehensive search
results = search_tool._run(
query="quantum computing",
depth="standard"
)
# Deep detailed search
results = search_tool._run(
query="quantum computing",
depth="deep"
)
```
### Output Type Options
```python
# Default search results
results = search_tool._run(
query="quantum computing",
output_type="searchResults"
)
# Custom output format
results = search_tool._run(
query="quantum computing",
output_type="customFormat"
)
```
### Error Handling
```python
results = search_tool._run(query="quantum computing")
if results["success"]:
for result in results["results"]:
print(f"Name: {result['name']}")
print(f"URL: {result['url']}")
print(f"Content: {result['content']}")
else:
print(f"Error: {results['error']}")
```
## Notes
- Requires valid Linkup API key
- Returns structured search results
- Supports multiple search depths
- Configurable output formats
- Built-in error handling
- Thread-safe operations
- Efficient for contextual searches

View File

@@ -0,0 +1,192 @@
---
title: LlamaIndexTool
description: A wrapper tool for integrating LlamaIndex tools and query engines with CrewAI
icon: link
---
## LlamaIndexTool
The LlamaIndexTool serves as a bridge between CrewAI and LlamaIndex, allowing you to use LlamaIndex tools and query engines within your CrewAI agents. It supports both direct tool wrapping and query engine integration.
## Installation
```bash
pip install 'crewai[tools]'
pip install llama-index # Required for LlamaIndex integration
```
## Usage Examples
### Using with LlamaIndex Tools
```python
from crewai import Agent
from crewai_tools import LlamaIndexTool
from llama_index.core.tools import BaseTool as LlamaBaseTool
from pydantic import BaseModel, Field
# Create a LlamaIndex tool
class CustomLlamaSchema(BaseModel):
query: str = Field(..., description="Query to process")
class CustomLlamaTool(LlamaBaseTool):
name = "Custom Llama Tool"
description = "A custom LlamaIndex tool"
def __call__(self, query: str) -> str:
return f"Processed: {query}"
# Wrap the LlamaIndex tool
llama_tool = CustomLlamaTool()
wrapped_tool = LlamaIndexTool.from_tool(llama_tool)
# Create an agent with the tool
agent = Agent(
role='LlamaIndex Integration Agent',
goal='Process queries using LlamaIndex tools',
backstory='Specialist in integrating LlamaIndex capabilities.',
tools=[wrapped_tool]
)
```
### Using with Query Engines
```python
from crewai import Agent
from crewai_tools import LlamaIndexTool
from llama_index.core import VectorStoreIndex, Document
# Create a query engine
documents = [Document(text="Sample document content")]
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()
# Create the tool
query_tool = LlamaIndexTool.from_query_engine(
query_engine,
name="Document Search",
description="Search through indexed documents"
)
# Create an agent with the tool
agent = Agent(
role='Document Researcher',
goal='Find relevant information in documents',
backstory='Expert at searching through document collections.',
tools=[query_tool]
)
```
## Tool Creation Methods
### From LlamaIndex Tool
```python
@classmethod
def from_tool(cls, tool: Any, **kwargs: Any) -> "LlamaIndexTool":
"""
Create a CrewAI tool from a LlamaIndex tool.
Args:
tool (LlamaBaseTool): A LlamaIndex tool to wrap
**kwargs: Additional arguments for tool creation
Returns:
LlamaIndexTool: A CrewAI-compatible tool wrapper
Raises:
ValueError: If tool is not a LlamaBaseTool or lacks fn_schema
"""
```
### From Query Engine
```python
@classmethod
def from_query_engine(
cls,
query_engine: Any,
name: Optional[str] = None,
description: Optional[str] = None,
return_direct: bool = False,
**kwargs: Any
) -> "LlamaIndexTool":
"""
Create a CrewAI tool from a LlamaIndex query engine.
Args:
query_engine (BaseQueryEngine): The query engine to wrap
name (Optional[str]): Custom name for the tool
description (Optional[str]): Custom description
return_direct (bool): Whether to return query engine response directly
**kwargs: Additional arguments for tool creation
Returns:
LlamaIndexTool: A CrewAI-compatible tool wrapper
Raises:
ValueError: If query_engine is not a BaseQueryEngine
"""
```
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import LlamaIndexTool
from llama_index.core import VectorStoreIndex, Document
from llama_index.core.tools import QueryEngineTool
# Create documents and index
documents = [
Document(text="AI is a technology that simulates human intelligence."),
Document(text="Machine learning is a subset of AI.")
]
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()
# Create the tool
search_tool = LlamaIndexTool.from_query_engine(
query_engine,
name="AI Knowledge Base",
description="Search through AI-related documents"
)
# Create agent
researcher = Agent(
role='AI Researcher',
goal='Research AI concepts',
backstory='Expert at finding and explaining AI concepts.',
tools=[search_tool]
)
# Define task
research_task = Task(
description="""Find and explain what AI is and its relationship
with machine learning.""",
agent=researcher
)
# The agent will use:
# {
# "query": "What is AI and how does it relate to machine learning?"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Notes
- Automatically adapts LlamaIndex tool schemas for CrewAI compatibility
- Renames 'input' parameter to 'query' for better integration
- Supports both direct tool wrapping and query engine integration
- Handles schema validation and error resolution
- Thread-safe operations
- Compatible with all LlamaIndex tool types and query engines

View File

@@ -0,0 +1,209 @@
---
title: MDX Search Tool
description: A tool for semantic searching within MDX files using RAG capabilities
---
# MDX Search Tool
The MDX Search Tool enables semantic searching within MDX (Markdown with JSX) files using Retrieval-Augmented Generation (RAG) capabilities. It supports both fixed and dynamic file path modes, allowing you to specify the MDX file at initialization or runtime.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage
You can use the MDX Search Tool in two ways:
### 1. Fixed MDX File Path
Initialize the tool with a specific MDX file path:
```python
from crewai import Agent
from crewai_tools import MDXSearchTool
# Initialize with a fixed MDX file
tool = MDXSearchTool(mdx="/path/to/your/document.mdx")
# Create an agent with the tool
agent = Agent(
role='Technical Writer',
goal='Search through MDX documentation',
backstory='I help find relevant information in MDX documentation',
tools=[tool]
)
# Use in a task
task = Task(
description="Find information about API endpoints in the documentation",
agent=agent
)
```
### 2. Dynamic MDX File Path
Initialize the tool without a specific file path to provide it at runtime:
```python
from crewai import Agent
from crewai_tools import MDXSearchTool
# Initialize without a fixed MDX file
tool = MDXSearchTool()
# Create an agent with the tool
agent = Agent(
role='Documentation Analyst',
goal='Search through various MDX files',
backstory='I analyze different MDX documentation files',
tools=[tool]
)
# Use in a task with dynamic file path
task = Task(
description="Search for 'authentication' in the API documentation",
agent=agent,
context={
"mdx": "/path/to/api-docs.mdx",
"search_query": "authentication"
}
)
```
## Input Schema
### Fixed MDX File Mode
```python
class FixedMDXSearchToolSchema(BaseModel):
search_query: str # The search query to find content in the MDX file
```
### Dynamic MDX File Mode
```python
class MDXSearchToolSchema(BaseModel):
search_query: str # The search query to find content in the MDX file
mdx: str # The path to the MDX file to search
```
## Function Signatures
```python
def __init__(self, mdx: Optional[str] = None, **kwargs):
"""
Initialize the MDX Search Tool.
Args:
mdx (Optional[str]): Path to the MDX file (optional)
**kwargs: Additional arguments passed to RagTool
"""
def _run(
self,
search_query: str,
**kwargs: Any,
) -> str:
"""
Execute the search on the MDX file.
Args:
search_query (str): The query to search for
**kwargs: Additional arguments including 'mdx' for dynamic mode
Returns:
str: The search results from the MDX content
"""
```
## Best Practices
1. **File Path Handling**:
- Use absolute paths to avoid path resolution issues
- Verify file existence before searching
- Handle file permissions appropriately
2. **Query Optimization**:
- Use specific, focused search queries
- Consider context when formulating queries
- Break down complex searches into smaller queries
3. **Error Handling**:
- Handle file not found errors gracefully
- Manage permission issues appropriately
- Validate input parameters before processing
## Example Integration
Here's a complete example showing how to integrate the MDX Search Tool with CrewAI:
```python
from crewai import Agent, Task, Crew
from crewai_tools import MDXSearchTool
# Initialize the tool
mdx_tool = MDXSearchTool()
# Create an agent with the tool
researcher = Agent(
role='Documentation Researcher',
goal='Find and analyze information in MDX documentation',
backstory='I am an expert at finding relevant information in documentation',
tools=[mdx_tool]
)
# Create tasks
search_task = Task(
description="""
Search through the API documentation for information about authentication methods.
Look for:
1. Authentication endpoints
2. Security best practices
3. Token handling
Provide a comprehensive summary of the findings.
""",
agent=researcher,
context={
"mdx": "/path/to/api-docs.mdx",
"search_query": "authentication security tokens"
}
)
# Create and run the crew
crew = Crew(
agents=[researcher],
tasks=[search_task]
)
result = crew.kickoff()
```
## Error Handling
The tool handles various error scenarios:
1. **File Not Found**:
```python
try:
tool = MDXSearchTool(mdx="/path/to/nonexistent.mdx")
except FileNotFoundError:
print("MDX file not found. Please verify the file path.")
```
2. **Permission Issues**:
```python
try:
tool = MDXSearchTool(mdx="/restricted/docs.mdx")
except PermissionError:
print("Insufficient permissions to access the MDX file.")
```
3. **Invalid Content**:
```python
try:
result = tool._run(search_query="query", mdx="/path/to/invalid.mdx")
except ValueError:
print("Invalid MDX content or format.")
```

View File

@@ -0,0 +1,217 @@
---
title: MySQLSearchTool
description: A tool for semantic search within MySQL database tables using RAG capabilities
icon: database
---
## MySQLSearchTool
The MySQLSearchTool enables semantic search capabilities for MySQL database tables using Retrieval-Augmented Generation (RAG). It processes table contents and allows natural language queries to search through the data.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import MySQLSearchTool
# Initialize the tool
mysql_tool = MySQLSearchTool(
table_name="users",
db_uri="mysql://user:pass@localhost:3306/database"
)
# Create an agent with the tool
researcher = Agent(
role='Database Researcher',
goal='Search and analyze database contents',
backstory='Expert at finding relevant information in databases.',
tools=[mysql_tool],
verbose=True
)
```
## Input Schema
```python
class MySQLSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory semantic search query you want to use to search the database's content"
)
```
## Function Signature
```python
def __init__(
self,
table_name: str,
db_uri: str,
**kwargs
):
"""
Initialize the MySQL search tool.
Args:
table_name (str): Name of the table to search
db_uri (str): Database connection URI
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on table contents.
Args:
search_query (str): Query to search in the table
**kwargs: Additional arguments
Returns:
str: Relevant content from the table matching the query
"""
```
## Best Practices
1. Database Connection:
- Use secure connection URIs
- Handle authentication properly
- Manage connection lifecycle
- Monitor timeouts
2. Query Optimization:
- Structure queries clearly
- Consider table size
- Handle large datasets
- Monitor performance
3. Security Considerations:
- Protect credentials
- Use environment variables
- Limit table access
- Validate inputs
4. Error Handling:
- Handle connection errors
- Manage query timeouts
- Provide clear messages
- Log issues
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import MySQLSearchTool
# Initialize tool
mysql_tool = MySQLSearchTool(
table_name="customers",
db_uri="mysql://user:pass@localhost:3306/crm"
)
# Create agent
researcher = Agent(
role='Database Analyst',
goal='Extract customer insights from database',
backstory='Expert at analyzing customer data.',
tools=[mysql_tool]
)
# Define task
analysis_task = Task(
description="""Find all premium customers
with recent purchases.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "premium customers recent purchases"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Multiple Table Analysis
```python
# Create tools for different tables
customers_tool = MySQLSearchTool(
table_name="customers",
db_uri="mysql://user:pass@localhost:3306/crm"
)
orders_tool = MySQLSearchTool(
table_name="orders",
db_uri="mysql://user:pass@localhost:3306/crm"
)
# Create agent with multiple tools
analyst = Agent(
role='Data Analyst',
goal='Cross-reference customer and order data',
tools=[customers_tool, orders_tool]
)
```
### Secure Connection Configuration
```python
import os
# Use environment variables for credentials
db_uri = (
f"mysql://{os.getenv('DB_USER')}:{os.getenv('DB_PASS')}"
f"@{os.getenv('DB_HOST')}:{os.getenv('DB_PORT')}"
f"/{os.getenv('DB_NAME')}"
)
tool = MySQLSearchTool(
table_name="sensitive_data",
db_uri=db_uri
)
```
### Error Handling Example
```python
try:
mysql_tool = MySQLSearchTool(
table_name="users",
db_uri="mysql://user:pass@localhost:3306/app"
)
results = mysql_tool.run(
search_query="active users in California"
)
print(results)
except Exception as e:
print(f"Error querying database: {str(e)}")
```
## Notes
- Inherits from RagTool
- Uses MySQLLoader
- Requires database URI
- Table-specific search
- Semantic query support
- Connection management
- Error handling
- Performance optimization
- Security features
- Memory efficiency

View File

@@ -0,0 +1,208 @@
---
title: PDFSearchTool
description: A tool for semantic search within PDF documents using RAG capabilities
icon: file-search
---
## PDFSearchTool
The PDFSearchTool enables semantic search capabilities for PDF documents using Retrieval-Augmented Generation (RAG). It leverages embedchain's PDFEmbedchainAdapter for efficient PDF processing and supports both fixed and dynamic PDF path specification.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import PDFSearchTool
# Method 1: Initialize with specific PDF
pdf_tool = PDFSearchTool(pdf="/path/to/document.pdf")
# Method 2: Initialize without PDF (specify at runtime)
flexible_pdf_tool = PDFSearchTool()
# Create an agent with the tool
researcher = Agent(
role='PDF Researcher',
goal='Search and analyze PDF documents',
backstory='Expert at finding relevant information in PDFs.',
tools=[pdf_tool],
verbose=True
)
```
## Input Schema
### Fixed PDF Schema (when PDF path provided during initialization)
```python
class FixedPDFSearchToolSchema(BaseModel):
query: str = Field(
description="Mandatory query you want to use to search the PDF's content"
)
```
### Flexible PDF Schema (when PDF path provided at runtime)
```python
class PDFSearchToolSchema(FixedPDFSearchToolSchema):
pdf: str = Field(
description="Mandatory pdf path you want to search"
)
```
## Function Signature
```python
def __init__(
self,
pdf: Optional[str] = None,
**kwargs
):
"""
Initialize the PDF search tool.
Args:
pdf (Optional[str]): Path to PDF file (optional)
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on PDF content.
Args:
query (str): Search query for the PDF
**kwargs: Additional arguments including pdf path if not initialized
Returns:
str: Relevant content from the PDF matching the query
"""
```
## Best Practices
1. PDF File Handling:
- Use absolute paths for reliability
- Verify PDF file existence
- Handle large PDFs appropriately
2. Search Optimization:
- Use specific, focused queries
- Consider document structure
- Test with sample queries first
3. Performance Considerations:
- Pre-initialize with PDF for repeated searches
- Handle large documents efficiently
- Monitor memory usage
4. Error Handling:
- Verify PDF file existence
- Handle malformed PDFs
- Manage file access permissions
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import PDFSearchTool
# Initialize tool with specific PDF
pdf_tool = PDFSearchTool(pdf="/path/to/research.pdf")
# Create agent
researcher = Agent(
role='PDF Researcher',
goal='Extract insights from research papers',
backstory='Expert at analyzing research documents.',
tools=[pdf_tool]
)
# Define task
research_task = Task(
description="""Find all mentions of machine learning
applications in healthcare from the PDF.""",
agent=researcher
)
# The tool will use:
# {
# "query": "machine learning applications healthcare"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Dynamic PDF Selection
```python
# Initialize without PDF
flexible_tool = PDFSearchTool()
# Search different PDFs
research_results = flexible_tool.run(
query="quantum computing",
pdf="/path/to/research.pdf"
)
report_results = flexible_tool.run(
query="financial metrics",
pdf="/path/to/report.pdf"
)
```
### Multiple PDF Analysis
```python
# Create tools for different PDFs
research_tool = PDFSearchTool(pdf="/path/to/research.pdf")
report_tool = PDFSearchTool(pdf="/path/to/report.pdf")
# Create agent with multiple tools
analyst = Agent(
role='Document Analyst',
goal='Cross-reference multiple documents',
tools=[research_tool, report_tool]
)
```
### Error Handling Example
```python
try:
pdf_tool = PDFSearchTool()
results = pdf_tool.run(
query="important findings",
pdf="/path/to/document.pdf"
)
print(results)
except Exception as e:
print(f"Error processing PDF: {str(e)}")
```
## Notes
- Inherits from RagTool
- Uses PDFEmbedchainAdapter
- Supports semantic search
- Dynamic PDF specification
- Efficient content retrieval
- Thread-safe operations
- Maintains search context
- Handles large documents
- Supports various PDF formats
- Memory-efficient processing

View File

@@ -0,0 +1,234 @@
---
title: PDFTextWritingTool
description: A tool for adding text to specific positions in PDF documents with custom font support
icon: file-pdf
---
## PDFTextWritingTool
The PDFTextWritingTool allows you to add text to specific positions in PDF documents with support for custom fonts, colors, and positioning. It's particularly useful for adding annotations, watermarks, or any text overlay to existing PDFs.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import PDFTextWritingTool
# Basic initialization
pdf_tool = PDFTextWritingTool()
# Create an agent with the tool
document_processor = Agent(
role='Document Processor',
goal='Add text annotations to PDF documents',
backstory='Expert at PDF document processing and text manipulation.',
tools=[pdf_tool],
verbose=True
)
```
## Input Schema
```python
class PDFTextWritingToolSchema(BaseModel):
pdf_path: str = Field(
description="Path to the PDF file to modify"
)
text: str = Field(
description="Text to add to the PDF"
)
position: tuple = Field(
description="Tuple of (x, y) coordinates for text placement"
)
font_size: int = Field(
default=12,
description="Font size of the text"
)
font_color: str = Field(
default="0 0 0 rg",
description="RGB color code for the text"
)
font_name: Optional[str] = Field(
default="F1",
description="Font name for standard fonts"
)
font_file: Optional[str] = Field(
default=None,
description="Path to a .ttf font file for custom font usage"
)
page_number: int = Field(
default=0,
description="Page number to add text to"
)
```
## Function Signature
```python
def run(
self,
pdf_path: str,
text: str,
position: tuple,
font_size: int,
font_color: str,
font_name: str = "F1",
font_file: Optional[str] = None,
page_number: int = 0,
**kwargs
) -> str:
"""
Add text to a specific position in a PDF document.
Args:
pdf_path (str): Path to the PDF file to modify
text (str): Text to add to the PDF
position (tuple): (x, y) coordinates for text placement
font_size (int): Font size of the text
font_color (str): RGB color code for the text (e.g., "0 0 0 rg" for black)
font_name (str, optional): Font name for standard fonts (default: "F1")
font_file (str, optional): Path to a .ttf font file for custom font
page_number (int, optional): Page number to add text to (default: 0)
Returns:
str: Success message with output file path
"""
```
## Best Practices
1. File Handling:
- Ensure PDF files exist before processing
- Use absolute paths for reliability
- Handle file permissions appropriately
2. Text Positioning:
- Use appropriate coordinates based on PDF dimensions
- Consider page orientation and margins
- Test positioning with small changes first
3. Font Usage:
- Verify custom font files exist
- Use standard fonts when possible
- Test font rendering before production use
4. Error Handling:
- Check page numbers are valid
- Verify font file accessibility
- Handle file writing permissions
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import PDFTextWritingTool
# Initialize tool
pdf_tool = PDFTextWritingTool()
# Create agent
document_processor = Agent(
role='Document Processor',
goal='Process and annotate PDF documents',
backstory='Expert at PDF manipulation and text placement.',
tools=[pdf_tool]
)
# Define task
annotation_task = Task(
description="""Add a watermark saying 'CONFIDENTIAL' to
the center of the first page of the document at
'/path/to/document.pdf'.""",
agent=document_processor
)
# The tool will use:
# {
# "pdf_path": "/path/to/document.pdf",
# "text": "CONFIDENTIAL",
# "position": (300, 400),
# "font_size": 24,
# "font_color": "1 0 0 rg", # Red color
# "page_number": 0
# }
# Create crew
crew = Crew(
agents=[document_processor],
tasks=[annotation_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Custom Font Example
```python
# Using a custom font
result = pdf_tool.run(
pdf_path="/path/to/input.pdf",
text="Custom Font Text",
position=(100, 500),
font_size=16,
font_color="0 0 1 rg", # Blue color
font_file="/path/to/custom_font.ttf",
page_number=0
)
```
### Multiple Text Elements
```python
# Add multiple text elements
positions = [(100, 700), (100, 650), (100, 600)]
texts = ["Header", "Subheader", "Body Text"]
font_sizes = [18, 14, 12]
for text, position, size in zip(texts, positions, font_sizes):
pdf_tool.run(
pdf_path="/path/to/input.pdf",
text=text,
position=position,
font_size=size,
font_color="0 0 0 rg" # Black color
)
```
### Color Text Example
```python
# Add colored text
colors = {
"red": "1 0 0 rg",
"green": "0 1 0 rg",
"blue": "0 0 1 rg"
}
for y_pos, (color_name, color_code) in enumerate(colors.items()):
pdf_tool.run(
pdf_path="/path/to/input.pdf",
text=f"This text is {color_name}",
position=(100, 700 - y_pos * 50),
font_size=14,
font_color=color_code
)
```
## Notes
- Supports custom TrueType fonts (.ttf)
- Allows RGB color specifications
- Handles multi-page PDFs
- Preserves original PDF content
- Supports text positioning with x,y coordinates
- Maintains PDF structure and metadata
- Creates new output file for safety
- Thread-safe operations
- Efficient PDF manipulation
- Supports various text attributes

View File

@@ -0,0 +1,181 @@
---
title: PGSearchTool
description: A RAG-based semantic search tool for PostgreSQL database content
icon: database-search
---
## PGSearchTool
The PGSearchTool provides semantic search capabilities for PostgreSQL database content using RAG (Retrieval-Augmented Generation). It allows for natural language queries over database table content by leveraging embeddings and semantic search.
## Installation
```bash
pip install 'crewai[tools]'
pip install embedchain # Required dependency
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import PGSearchTool
# Initialize the tool with database configuration
search_tool = PGSearchTool(
db_uri="postgresql://user:password@localhost:5432/dbname",
table_name="your_table"
)
# Create an agent with the tool
researcher = Agent(
role='Database Researcher',
goal='Find relevant information in database content',
backstory='Expert at searching and analyzing database content.',
tools=[search_tool],
verbose=True
)
```
## Input Schema
```python
class PGSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory semantic search query for searching the database's content"
)
```
## Function Signature
```python
def __init__(self, table_name: str, **kwargs):
"""
Initialize the PostgreSQL search tool.
Args:
table_name (str): Name of the table to search
db_uri (str): PostgreSQL database URI (required in kwargs)
**kwargs: Additional arguments for RagTool initialization
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> Any:
"""
Perform semantic search on database content.
Args:
search_query (str): Semantic search query
**kwargs: Additional search parameters
Returns:
Any: Relevant database content based on semantic search
"""
```
## Best Practices
1. Secure database credentials:
```python
# Use environment variables for sensitive data
import os
db_uri = (
f"postgresql://{os.getenv('DB_USER')}:{os.getenv('DB_PASS')}"
f"@{os.getenv('DB_HOST')}:{os.getenv('DB_PORT')}/{os.getenv('DB_NAME')}"
)
```
2. Optimize table selection
3. Use specific semantic queries
4. Handle database connection errors
5. Consider table size and query performance
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import PGSearchTool
# Initialize tool with database configuration
db_search = PGSearchTool(
db_uri="postgresql://user:password@localhost:5432/dbname",
table_name="customer_feedback"
)
# Create agent
analyst = Agent(
role='Database Analyst',
goal='Analyze customer feedback data',
backstory='Expert at finding insights in customer feedback.',
tools=[db_search]
)
# Define task
analysis_task = Task(
description="""Find all customer feedback related to product usability
and ease of use. Focus on common patterns and issues.""",
agent=analyst
)
# The tool will use:
# {
# "search_query": "product usability feedback ease of use issues"
# }
# Create crew
crew = Crew(
agents=[analyst],
tasks=[analysis_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Multiple Table Search
```python
# Create tools for different tables
customer_search = PGSearchTool(
db_uri="postgresql://user:password@localhost:5432/dbname",
table_name="customers"
)
orders_search = PGSearchTool(
db_uri="postgresql://user:password@localhost:5432/dbname",
table_name="orders"
)
# Use both tools in an agent
analyst = Agent(
role='Multi-table Analyst',
goal='Analyze customer and order data',
tools=[customer_search, orders_search]
)
```
### Error Handling
```python
try:
results = search_tool._run(
search_query="customer satisfaction ratings"
)
# Process results
except Exception as e:
print(f"Database search error: {str(e)}")
```
## Notes
- Inherits from RagTool for semantic search
- Uses embedchain's PostgresLoader
- Requires valid PostgreSQL connection
- Supports semantic natural language queries
- Thread-safe operations
- Efficient for large tables
- Handles connection pooling automatically

282
docs/tools/rag-tool.mdx Normal file
View File

@@ -0,0 +1,282 @@
---
title: RagTool
description: Base class for Retrieval-Augmented Generation (RAG) tools with flexible adapter support
icon: database
---
## RagTool
The RagTool serves as the base class for all Retrieval-Augmented Generation (RAG) tools in the CrewAI ecosystem. It provides a flexible adapter-based architecture for implementing knowledge base functionality with semantic search capabilities.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import RagTool
from crewai_tools.adapters import EmbedchainAdapter
from embedchain import App
# Create custom adapter
class CustomAdapter(RagTool.Adapter):
def query(self, question: str) -> str:
# Implement custom query logic
return "Answer based on knowledge base"
def add(self, *args, **kwargs) -> None:
# Implement custom add logic
pass
# Method 1: Use default EmbedchainAdapter
rag_tool = RagTool(
name="Custom Knowledge Base",
description="Specialized knowledge base for domain data",
summarize=True
)
# Method 2: Use custom adapter
custom_tool = RagTool(
name="Custom Knowledge Base",
adapter=CustomAdapter(),
summarize=False
)
# Create an agent with the tool
researcher = Agent(
role='Knowledge Base Researcher',
goal='Search and analyze knowledge base content',
backstory='Expert at finding relevant information in specialized datasets.',
tools=[rag_tool],
verbose=True
)
```
## Adapter Interface
```python
class Adapter(BaseModel, ABC):
@abstractmethod
def query(self, question: str) -> str:
"""
Query the knowledge base with a question.
Args:
question (str): Query to search in knowledge base
Returns:
str: Answer based on knowledge base content
"""
@abstractmethod
def add(self, *args: Any, **kwargs: Any) -> None:
"""
Add content to the knowledge base.
Args:
*args: Variable length argument list
**kwargs: Arbitrary keyword arguments
"""
```
## Function Signature
```python
def __init__(
self,
name: str = "Knowledge base",
description: str = "A knowledge base that can be used to answer questions.",
summarize: bool = False,
adapter: Optional[Adapter] = None,
config: Optional[dict[str, Any]] = None,
**kwargs
):
"""
Initialize the RAG tool.
Args:
name (str): Tool name
description (str): Tool description
summarize (bool): Enable answer summarization
adapter (Optional[Adapter]): Custom adapter implementation
config (Optional[dict]): Configuration for default adapter
**kwargs: Additional arguments for base tool
"""
def _run(
self,
query: str,
**kwargs: Any
) -> str:
"""
Execute query against knowledge base.
Args:
query (str): Question to ask
**kwargs: Additional arguments
Returns:
str: Answer from knowledge base
"""
```
## Best Practices
1. Adapter Implementation:
- Define clear interfaces
- Handle edge cases
- Implement error handling
- Document behavior
2. Knowledge Base Management:
- Organize content logically
- Update content regularly
- Monitor performance
- Handle large datasets
3. Query Optimization:
- Structure queries clearly
- Consider context
- Handle ambiguity
- Validate inputs
4. Error Handling:
- Handle missing data
- Manage timeouts
- Provide clear messages
- Log issues
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import RagTool
from embedchain import App
# Initialize tool with custom configuration
rag_tool = RagTool(
name="Technical Documentation KB",
description="Knowledge base for technical documentation",
summarize=True,
config={
"collection_name": "tech_docs",
"chunking": {
"chunk_size": 500,
"chunk_overlap": 50
}
}
)
# Add content to knowledge base
rag_tool.add(
"Technical documentation content here...",
data_type="text"
)
# Create agent
researcher = Agent(
role='Documentation Expert',
goal='Extract technical information from documentation',
backstory='Expert at analyzing technical documentation.',
tools=[rag_tool]
)
# Define task
research_task = Task(
description="""Find all mentions of API endpoints
and their authentication requirements.""",
agent=researcher
)
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Custom Adapter Implementation
```python
from typing import Any
from pydantic import BaseModel
from abc import ABC, abstractmethod
class SpecializedAdapter(RagTool.Adapter):
def __init__(self, config: dict):
self.config = config
self.knowledge_base = {}
def query(self, question: str) -> str:
# Implement specialized query logic
return self._process_query(question)
def add(self, content: str, **kwargs: Any) -> None:
# Implement specialized content addition
self._process_content(content, **kwargs)
# Use custom adapter
specialized_tool = RagTool(
name="Specialized KB",
adapter=SpecializedAdapter(config={"mode": "advanced"})
)
```
### Configuration Management
```python
# Configure default EmbedchainAdapter
config = {
"collection_name": "custom_collection",
"embedding": {
"model": "sentence-transformers/all-mpnet-base-v2",
"dimensions": 768
},
"chunking": {
"chunk_size": 1000,
"chunk_overlap": 100
}
}
tool = RagTool(config=config)
```
### Error Handling Example
```python
try:
rag_tool = RagTool()
# Add content
rag_tool.add(
"Documentation content...",
data_type="text"
)
# Query content
result = rag_tool.run(
query="What are the system requirements?"
)
print(result)
except Exception as e:
print(f"Error using knowledge base: {str(e)}")
```
## Notes
- Base class for RAG tools
- Flexible adapter pattern
- Default EmbedchainAdapter
- Custom adapter support
- Content management
- Query processing
- Error handling
- Configuration options
- Performance optimization
- Memory management

View File

@@ -0,0 +1,229 @@
---
title: SerpApi Google Search Tool
description: A tool for performing Google searches using the SerpApi service
---
# SerpApi Google Search Tool
The SerpApi Google Search Tool enables performing Google searches using the SerpApi service. It provides location-aware search capabilities with comprehensive result filtering.
## Installation
```bash
pip install 'crewai[tools]'
pip install serpapi
```
## Prerequisites
You need a SerpApi API key to use this tool. You can get one from [SerpApi's website](https://serpapi.com/manage-api-key).
Set your API key as an environment variable:
```bash
export SERPAPI_API_KEY="your_api_key_here"
```
## Usage
Here's how to use the SerpApi Google Search Tool:
```python
from crewai import Agent
from crewai_tools import SerpApiGoogleSearchTool
# Initialize the tool
search_tool = SerpApiGoogleSearchTool()
# Create an agent with the tool
search_agent = Agent(
role='Web Researcher',
goal='Find accurate information online',
backstory='I help research and analyze online information',
tools=[search_tool]
)
# Use in a task
task = Task(
description="Research recent AI developments",
agent=search_agent,
context={
"search_query": "latest artificial intelligence breakthroughs 2024",
"location": "United States" # Optional
}
)
```
## Input Schema
```python
class SerpApiGoogleSearchToolSchema(BaseModel):
search_query: str # The search query for Google Search
location: Optional[str] = None # Optional location for localized results
```
## Function Signatures
### Base Tool Initialization
```python
def __init__(self, **kwargs):
"""
Initialize the SerpApi tool with API credentials.
Raises:
ImportError: If serpapi package is not installed
ValueError: If SERPAPI_API_KEY environment variable is not set
"""
```
### Search Execution
```python
def _run(
self,
**kwargs: Any,
) -> dict:
"""
Execute the Google search.
Args:
search_query (str): The search query
location (Optional[str]): Optional location for results
Returns:
dict: Filtered search results from Google
Raises:
HTTPError: If the API request fails
"""
```
## Best Practices
1. **API Key Management**:
- Store the API key securely in environment variables
- Never hardcode the API key in your code
- Verify API key validity before making requests
2. **Search Optimization**:
- Use specific, targeted search queries
- Include relevant keywords and time frames
- Leverage location parameter for regional results
3. **Error Handling**:
- Handle API rate limits gracefully
- Implement retry logic for failed requests
- Validate input parameters before making requests
## Example Integration
Here's a complete example showing how to integrate the SerpApi Google Search Tool with CrewAI:
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerpApiGoogleSearchTool
# Initialize the tool
search_tool = SerpApiGoogleSearchTool()
# Create an agent with the tool
researcher = Agent(
role='Research Analyst',
goal='Find and analyze current information',
backstory="""I am an expert at finding and analyzing
information from various online sources.""",
tools=[search_tool]
)
# Create tasks
research_task = Task(
description="""
Research the following topic:
1. Latest developments in quantum computing
2. Focus on practical applications
3. Include major company announcements
Provide a comprehensive analysis of the findings.
""",
agent=researcher,
context={
"search_query": "quantum computing breakthroughs applications companies",
"location": "United States"
}
)
# Create and run the crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
result = crew.kickoff()
```
## Error Handling
The tool handles various error scenarios:
1. **Missing API Key**:
```python
try:
tool = SerpApiGoogleSearchTool()
except ValueError as e:
print("API key not found. Set SERPAPI_API_KEY environment variable.")
```
2. **API Request Errors**:
```python
try:
results = tool._run(
search_query="quantum computing",
location="United States"
)
except HTTPError as e:
print(f"API request failed: {str(e)}")
```
3. **Invalid Parameters**:
```python
try:
results = tool._run(
search_query="", # Empty query
location="Invalid Location"
)
except ValueError as e:
print("Invalid search parameters provided.")
```
## Response Format
The tool returns a filtered dictionary containing Google search results. Example response structure:
```python
{
"organic_results": [
{
"title": "Page Title",
"link": "https://...",
"snippet": "Page description or excerpt...",
"position": 1
}
# Additional results...
],
"knowledge_graph": {
"title": "Topic Title",
"description": "Topic description...",
"source": {
"name": "Source Name",
"link": "https://..."
}
},
"related_questions": [
{
"question": "Related question?",
"answer": "Answer to related question..."
}
# Additional related questions...
]
}
```
The response is automatically filtered to remove metadata and unnecessary fields, focusing on the most relevant search information. Fields like search metadata, parameters, and pagination are omitted for clarity.

View File

@@ -0,0 +1,225 @@
---
title: SerpApi Google Shopping Tool
description: A tool for searching Google Shopping using the SerpApi service
---
# SerpApi Google Shopping Tool
The SerpApi Google Shopping Tool enables searching Google Shopping results using the SerpApi service. It provides location-aware shopping search capabilities with comprehensive result filtering.
## Installation
```bash
pip install 'crewai[tools]'
pip install serpapi
```
## Prerequisites
You need a SerpApi API key to use this tool. You can get one from [SerpApi's website](https://serpapi.com/manage-api-key).
Set your API key as an environment variable:
```bash
export SERPAPI_API_KEY="your_api_key_here"
```
## Usage
Here's how to use the SerpApi Google Shopping Tool:
```python
from crewai import Agent
from crewai_tools import SerpApiGoogleShoppingTool
# Initialize the tool
shopping_tool = SerpApiGoogleShoppingTool()
# Create an agent with the tool
shopping_agent = Agent(
role='Shopping Researcher',
goal='Find the best shopping deals',
backstory='I help find and analyze shopping options',
tools=[shopping_tool]
)
# Use in a task
task = Task(
description="Find best deals for gaming laptops",
agent=shopping_agent,
context={
"search_query": "gaming laptop deals",
"location": "United States" # Optional
}
)
```
## Input Schema
```python
class SerpApiGoogleShoppingToolSchema(BaseModel):
search_query: str # The search query for Google Shopping
location: Optional[str] = None # Optional location for localized results
```
## Function Signatures
### Base Tool Initialization
```python
def __init__(self, **kwargs):
"""
Initialize the SerpApi tool with API credentials.
Raises:
ImportError: If serpapi package is not installed
ValueError: If SERPAPI_API_KEY environment variable is not set
"""
```
### Search Execution
```python
def _run(
self,
**kwargs: Any,
) -> dict:
"""
Execute the Google Shopping search.
Args:
search_query (str): The search query for Google Shopping
location (Optional[str]): Optional location for results
Returns:
dict: Filtered search results from Google Shopping
Raises:
HTTPError: If the API request fails
"""
```
## Best Practices
1. **API Key Management**:
- Store the API key securely in environment variables
- Never hardcode the API key in your code
- Verify API key validity before making requests
2. **Search Optimization**:
- Use specific, targeted search queries
- Include relevant product details in queries
- Leverage location parameter for regional pricing
3. **Error Handling**:
- Handle API rate limits gracefully
- Implement retry logic for failed requests
- Validate input parameters before making requests
## Example Integration
Here's a complete example showing how to integrate the SerpApi Google Shopping Tool with CrewAI:
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerpApiGoogleShoppingTool
# Initialize the tool
shopping_tool = SerpApiGoogleShoppingTool()
# Create an agent with the tool
researcher = Agent(
role='Shopping Analyst',
goal='Find and analyze the best shopping deals',
backstory="""I am an expert at finding the best shopping deals
and analyzing product offerings across different regions.""",
tools=[shopping_tool]
)
# Create tasks
search_task = Task(
description="""
Research gaming laptops with the following criteria:
1. Price range: $800-$1500
2. Released in the last year
3. Compare prices across different retailers
Provide a comprehensive analysis of the findings.
""",
agent=researcher,
context={
"search_query": "gaming laptop RTX 4060 2023",
"location": "United States"
}
)
# Create and run the crew
crew = Crew(
agents=[researcher],
tasks=[search_task]
)
result = crew.kickoff()
```
## Error Handling
The tool handles various error scenarios:
1. **Missing API Key**:
```python
try:
tool = SerpApiGoogleShoppingTool()
except ValueError as e:
print("API key not found. Set SERPAPI_API_KEY environment variable.")
```
2. **API Request Errors**:
```python
try:
results = tool._run(
search_query="gaming laptop",
location="United States"
)
except HTTPError as e:
print(f"API request failed: {str(e)}")
```
3. **Invalid Parameters**:
```python
try:
results = tool._run(
search_query="", # Empty query
location="Invalid Location"
)
except ValueError as e:
print("Invalid search parameters provided.")
```
## Response Format
The tool returns a filtered dictionary containing Google Shopping results. Example response structure:
```python
{
"shopping_results": [
{
"title": "Product Title",
"price": "$999.99",
"link": "https://...",
"source": "Retailer Name",
"rating": 4.5,
"reviews": 123,
"thumbnail": "https://..."
}
# Additional results...
],
"organic_results": [
{
"title": "Related Product",
"link": "https://...",
"snippet": "Product description..."
}
# Additional organic results...
]
}
```
The response is automatically filtered to remove metadata and unnecessary fields, focusing on the most relevant shopping information.

View File

@@ -0,0 +1,184 @@
---
title: SerplyJobSearchTool
description: A tool for searching US job postings using the Serply API
icon: briefcase
---
## SerplyJobSearchTool
The SerplyJobSearchTool provides job search capabilities using the Serply API. It allows for searching job postings in the US market, returning structured information about positions, employers, locations, and remote work status.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import SerplyJobSearchTool
# Set environment variable
# export SERPLY_API_KEY='your-api-key'
# Initialize the tool
search_tool = SerplyJobSearchTool()
# Create an agent with the tool
job_researcher = Agent(
role='Job Market Researcher',
goal='Find relevant job opportunities',
backstory='Expert at analyzing job market trends and opportunities.',
tools=[search_tool],
verbose=True
)
```
## Input Schema
```python
class SerplyJobSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query for fetching job postings"
)
```
## Function Signature
```python
def __init__(self, **kwargs):
"""
Initialize the job search tool.
Args:
**kwargs: Additional arguments for RagTool initialization
Note:
Requires SERPLY_API_KEY environment variable
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Perform job search using Serply API.
Args:
search_query (str): Job search query
**kwargs: Additional search parameters
Returns:
str: Formatted string containing job listings with details:
- Position
- Employer
- Location
- Link
- Highlights
- Remote/Hybrid status
"""
```
## Best Practices
1. Set up API authentication:
```bash
export SERPLY_API_KEY='your-serply-api-key'
```
2. Use specific search queries
3. Handle potential API errors
4. Process structured results effectively
5. Consider rate limits and quotas
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerplyJobSearchTool
# Initialize tool
job_search = SerplyJobSearchTool()
# Create agent
recruiter = Agent(
role='Technical Recruiter',
goal='Find relevant job opportunities in tech',
backstory='Expert at identifying promising tech positions.',
tools=[job_search]
)
# Define task
search_task = Task(
description="""Search for senior software engineer positions
with remote work options in the US. Focus on positions
requiring Python expertise.""",
agent=recruiter
)
# The tool will use:
# {
# "search_query": "senior software engineer python remote"
# }
# Create crew
crew = Crew(
agents=[recruiter],
tasks=[search_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Handling Search Results
```python
# Example of processing structured results
results = search_tool._run(
search_query="machine learning engineer"
)
# Results format:
"""
Search results:
Position: Senior Machine Learning Engineer
Employer: TechCorp Inc
Location: San Francisco, CA
Link: https://example.com/job/123
Highlights: Python, TensorFlow, 5+ years experience
Is Remote: True
Is Hybrid: False
---
Position: ML Engineer
...
"""
```
### Error Handling
```python
try:
results = search_tool._run(
search_query="data scientist"
)
if not results:
print("No jobs found")
else:
print(results)
except Exception as e:
print(f"Job search error: {str(e)}")
```
## Notes
- Requires valid Serply API key
- Currently supports US job market only
- Returns structured job information
- Includes remote/hybrid status
- Thread-safe operations
- Efficient job search capabilities
- Handles API rate limiting automatically
- Provides detailed job highlights

View File

@@ -0,0 +1,209 @@
---
title: SerplyNewsSearchTool
description: A news article search tool powered by Serply API with configurable search parameters
icon: newspaper
---
## SerplyNewsSearchTool
The SerplyNewsSearchTool provides news article search capabilities using the Serply API. It allows for customizable search parameters including result limits and proxy location for region-specific news results.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import SerplyNewsSearchTool
# Set environment variable
# export SERPLY_API_KEY='your-api-key'
# Basic initialization
news_tool = SerplyNewsSearchTool()
# Advanced initialization with custom parameters
news_tool = SerplyNewsSearchTool(
limit=20, # Return 20 results
proxy_location="FR" # Search from France
)
# Create an agent with the tool
news_researcher = Agent(
role='News Researcher',
goal='Find relevant news articles',
backstory='Expert at news research and information gathering.',
tools=[news_tool],
verbose=True
)
```
## Input Schema
```python
class SerplyNewsSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query for fetching news articles"
)
```
## Function Signature
```python
def __init__(
self,
limit: Optional[int] = 10,
proxy_location: Optional[str] = "US",
**kwargs
):
"""
Initialize the news search tool.
Args:
limit (int): Maximum number of results [10-100] (default: 10)
proxy_location (str): Region for local news results (default: "US")
Options: US, CA, IE, GB, FR, DE, SE, IN, JP, KR, SG, AU, BR
**kwargs: Additional arguments for tool creation
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Perform news search using Serply API.
Args:
search_query (str): News search query
Returns:
str: Formatted string containing news results:
- Title
- Link
- Source
- Published Date
"""
```
## Best Practices
1. Set up API authentication:
```bash
export SERPLY_API_KEY='your-serply-api-key'
```
2. Configure search parameters appropriately:
- Set reasonable result limits
- Select relevant proxy location for regional news
- Consider time sensitivity of news content
3. Handle potential API errors
4. Process structured results effectively
5. Consider rate limits and quotas
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerplyNewsSearchTool
# Initialize tool with custom configuration
news_tool = SerplyNewsSearchTool(
limit=15, # 15 results
proxy_location="US" # US news sources
)
# Create agent
news_analyst = Agent(
role='News Analyst',
goal='Research breaking news and developments',
backstory='Expert at analyzing news trends and developments.',
tools=[news_tool]
)
# Define task
news_task = Task(
description="""Research the latest developments in renewable
energy technology and investments, focusing on major
announcements and industry trends.""",
agent=news_analyst
)
# The tool will use:
# {
# "search_query": "renewable energy technology investments news"
# }
# Create crew
crew = Crew(
agents=[news_analyst],
tasks=[news_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Regional News Configuration
```python
# French news sources
fr_news = SerplyNewsSearchTool(
proxy_location="FR",
limit=20
)
# Japanese news sources
jp_news = SerplyNewsSearchTool(
proxy_location="JP",
limit=20
)
```
### Result Processing
```python
# Get news results
try:
results = news_tool._run(
search_query="renewable energy investments"
)
print(results)
except Exception as e:
print(f"News search error: {str(e)}")
```
### Multiple Region Search
```python
# Search across multiple regions
regions = ["US", "GB", "DE"]
all_results = []
for region in regions:
regional_tool = SerplyNewsSearchTool(
proxy_location=region,
limit=5
)
results = regional_tool._run(
search_query="global tech innovations"
)
all_results.append(f"Results from {region}:\n{results}")
combined_results = "\n\n".join(all_results)
```
## Notes
- Requires valid Serply API key
- Supports multiple regions for news sources
- Configurable result limits (10-100)
- Returns structured news article data
- Thread-safe operations
- Efficient news search capabilities
- Handles API rate limiting automatically
- Includes source attribution and publication dates
- Follows redirects for final article URLs

View File

@@ -0,0 +1,209 @@
---
title: SerplyScholarSearchTool
description: A scholarly literature search tool powered by Serply API with configurable search parameters
icon: book
---
## SerplyScholarSearchTool
The SerplyScholarSearchTool provides scholarly literature search capabilities using the Serply API. It allows for customizable search parameters including language and proxy location for region-specific academic results.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import SerplyScholarSearchTool
# Set environment variable
# export SERPLY_API_KEY='your-api-key'
# Basic initialization
scholar_tool = SerplyScholarSearchTool()
# Advanced initialization with custom parameters
scholar_tool = SerplyScholarSearchTool(
hl="fr", # French language results
proxy_location="FR" # Search from France
)
# Create an agent with the tool
academic_researcher = Agent(
role='Academic Researcher',
goal='Find relevant scholarly literature',
backstory='Expert at academic research and literature review.',
tools=[scholar_tool],
verbose=True
)
```
## Input Schema
```python
class SerplyScholarSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query for fetching scholarly literature"
)
```
## Function Signature
```python
def __init__(
self,
hl: str = "us",
proxy_location: Optional[str] = "US",
**kwargs
):
"""
Initialize the scholar search tool.
Args:
hl (str): Host language code for results (default: "us")
Reference: https://developers.google.com/custom-search/docs/xml_results?hl=en#wsInterfaceLanguages
proxy_location (str): Region for local results (default: "US")
Options: US, CA, IE, GB, FR, DE, SE, IN, JP, KR, SG, AU, BR
**kwargs: Additional arguments for tool creation
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Perform scholarly literature search using Serply API.
Args:
search_query (str): Academic search query
Returns:
str: Formatted string containing scholarly results:
- Title
- Link
- Description
- Citation
- Authors
"""
```
## Best Practices
1. Set up API authentication:
```bash
export SERPLY_API_KEY='your-serply-api-key'
```
2. Configure search parameters appropriately:
- Use relevant language codes
- Select appropriate proxy location
- Provide specific academic search terms
3. Handle potential API errors
4. Process structured results effectively
5. Consider rate limits and quotas
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerplyScholarSearchTool
# Initialize tool with custom configuration
scholar_tool = SerplyScholarSearchTool(
hl="en", # English results
proxy_location="US" # US academic sources
)
# Create agent
researcher = Agent(
role='Academic Researcher',
goal='Research recent academic publications',
backstory='Expert at analyzing academic literature and research trends.',
tools=[scholar_tool]
)
# Define task
research_task = Task(
description="""Research recent academic publications on
machine learning applications in healthcare, focusing on
peer-reviewed articles from the last two years.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "machine learning healthcare applications"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Language and Region Configuration
```python
# French academic sources
fr_scholar = SerplyScholarSearchTool(
hl="fr",
proxy_location="FR"
)
# German academic sources
de_scholar = SerplyScholarSearchTool(
hl="de",
proxy_location="DE"
)
```
### Result Processing
```python
try:
results = scholar_tool._run(
search_query="machine learning healthcare applications"
)
print(results)
except Exception as e:
print(f"Scholar search error: {str(e)}")
```
### Citation Analysis
```python
# Extract and analyze citations
def analyze_citations(results):
citations = []
for result in results.split("---"):
if "Cite:" in result:
citation = result.split("Cite:")[1].split("\n")[0].strip()
citations.append(citation)
return citations
results = scholar_tool._run(
search_query="artificial intelligence ethics"
)
citations = analyze_citations(results)
```
## Notes
- Requires valid Serply API key
- Supports multiple languages and regions
- Returns structured academic article data
- Includes citation information
- Lists all authors of publications
- Thread-safe operations
- Efficient scholarly search capabilities
- Handles API rate limiting automatically
- Supports both direct and document links
- Provides comprehensive article metadata

View File

@@ -0,0 +1,213 @@
---
title: SerplyWebSearchTool
description: A Google search tool powered by Serply API with configurable search parameters
icon: search
---
## SerplyWebSearchTool
The SerplyWebSearchTool provides Google search capabilities using the Serply API. It allows for customizable search parameters including language, result limits, device type, and proxy location for region-specific results.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import SerplyWebSearchTool
# Set environment variable
# export SERPLY_API_KEY='your-api-key'
# Basic initialization
search_tool = SerplyWebSearchTool()
# Advanced initialization with custom parameters
search_tool = SerplyWebSearchTool(
hl="fr", # French language results
limit=20, # Return 20 results
device_type="mobile", # Mobile search results
proxy_location="FR" # Search from France
)
# Create an agent with the tool
researcher = Agent(
role='Web Researcher',
goal='Find relevant information online',
backstory='Expert at web research and information gathering.',
tools=[search_tool],
verbose=True
)
```
## Input Schema
```python
class SerplyWebSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query for Google search"
)
```
## Function Signature
```python
def __init__(
self,
hl: str = "us",
limit: int = 10,
device_type: str = "desktop",
proxy_location: str = "US",
**kwargs
):
"""
Initialize the Google search tool.
Args:
hl (str): Host language code for results (default: "us")
Reference: https://developers.google.com/custom-search/docs/xml_results?hl=en#wsInterfaceLanguages
limit (int): Maximum number of results [10-100] (default: 10)
device_type (str): "desktop" or "mobile" results (default: "desktop")
proxy_location (str): Region for local results (default: "US")
Options: US, CA, IE, GB, FR, DE, SE, IN, JP, KR, SG, AU, BR
**kwargs: Additional arguments for tool creation
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Perform Google search using Serply API.
Args:
search_query (str): Search query
Returns:
str: Formatted string containing search results:
- Title
- Link
- Description
"""
```
## Best Practices
1. Set up API authentication:
```bash
export SERPLY_API_KEY='your-serply-api-key'
```
2. Configure search parameters appropriately:
- Use relevant language codes
- Set reasonable result limits
- Choose appropriate device type
- Select relevant proxy location
3. Handle potential API errors
4. Process structured results effectively
5. Consider rate limits and quotas
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerplyWebSearchTool
# Initialize tool with custom configuration
search_tool = SerplyWebSearchTool(
hl="en", # English results
limit=15, # 15 results
device_type="desktop",
proxy_location="US"
)
# Create agent
researcher = Agent(
role='Web Researcher',
goal='Research emerging technology trends',
backstory='Expert at finding and analyzing tech trends.',
tools=[search_tool]
)
# Define task
research_task = Task(
description="""Research the latest developments in artificial
intelligence and machine learning, focusing on practical
applications in business.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "latest AI ML developments business applications"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Language and Region Configuration
```python
# French search from France
fr_search = SerplyWebSearchTool(
hl="fr",
proxy_location="FR"
)
# Japanese search from Japan
jp_search = SerplyWebSearchTool(
hl="ja",
proxy_location="JP"
)
```
### Device-Specific Results
```python
# Mobile results
mobile_search = SerplyWebSearchTool(
device_type="mobile",
limit=20
)
# Desktop results
desktop_search = SerplyWebSearchTool(
device_type="desktop",
limit=20
)
```
### Error Handling
```python
try:
results = search_tool._run(
search_query="artificial intelligence trends"
)
print(results)
except Exception as e:
print(f"Search error: {str(e)}")
```
## Notes
- Requires valid Serply API key
- Supports multiple languages and regions
- Configurable result limits (10-100)
- Device-specific search results
- Thread-safe operations
- Efficient search capabilities
- Handles API rate limiting automatically
- Returns structured search results

View File

@@ -0,0 +1,201 @@
---
title: SerplyWebpageToMarkdownTool
description: A tool for converting web pages to markdown format using Serply API
icon: markdown
---
## SerplyWebpageToMarkdownTool
The SerplyWebpageToMarkdownTool converts web pages to markdown format using the Serply API, making it easier for LLMs to process and understand web content. It supports configurable proxy locations for region-specific access.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import SerplyWebpageToMarkdownTool
# Set environment variable
# export SERPLY_API_KEY='your-api-key'
# Basic initialization
markdown_tool = SerplyWebpageToMarkdownTool()
# Advanced initialization with custom parameters
markdown_tool = SerplyWebpageToMarkdownTool(
proxy_location="FR" # Access from France
)
# Create an agent with the tool
web_processor = Agent(
role='Web Content Processor',
goal='Convert web content to markdown format',
backstory='Expert at processing and formatting web content.',
tools=[markdown_tool],
verbose=True
)
```
## Input Schema
```python
class SerplyWebpageToMarkdownToolSchema(BaseModel):
url: str = Field(
description="Mandatory URL of the webpage to convert to markdown"
)
```
## Function Signature
```python
def __init__(
self,
proxy_location: Optional[str] = "US",
**kwargs
):
"""
Initialize the webpage to markdown conversion tool.
Args:
proxy_location (str): Region for accessing the webpage (default: "US")
Options: US, CA, IE, GB, FR, DE, SE, IN, JP, KR, SG, AU, BR
**kwargs: Additional arguments for tool creation
"""
def _run(
self,
**kwargs: Any
) -> str:
"""
Convert webpage to markdown using Serply API.
Args:
url (str): URL of the webpage to convert
Returns:
str: Markdown formatted content of the webpage
"""
```
## Best Practices
1. Set up API authentication:
```bash
export SERPLY_API_KEY='your-serply-api-key'
```
2. Configure proxy location appropriately:
- Select relevant region for access
- Consider content accessibility
- Handle region-specific content
3. Handle potential API errors
4. Process markdown output effectively
5. Consider rate limits and quotas
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import SerplyWebpageToMarkdownTool
# Initialize tool with custom configuration
markdown_tool = SerplyWebpageToMarkdownTool(
proxy_location="US" # US access point
)
# Create agent
processor = Agent(
role='Content Processor',
goal='Convert web content to structured markdown',
backstory='Expert at processing web content into structured formats.',
tools=[markdown_tool]
)
# Define task
conversion_task = Task(
description="""Convert the documentation page at
https://example.com/docs into markdown format for
further processing.""",
agent=processor
)
# The tool will use:
# {
# "url": "https://example.com/docs"
# }
# Create crew
crew = Crew(
agents=[processor],
tasks=[conversion_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Regional Access Configuration
```python
# European access points
fr_processor = SerplyWebpageToMarkdownTool(
proxy_location="FR"
)
de_processor = SerplyWebpageToMarkdownTool(
proxy_location="DE"
)
```
### Error Handling
```python
try:
markdown_content = markdown_tool._run(
url="https://example.com/page"
)
print(markdown_content)
except Exception as e:
print(f"Conversion error: {str(e)}")
```
### Content Processing
```python
# Process multiple pages
urls = [
"https://example.com/page1",
"https://example.com/page2",
"https://example.com/page3"
]
markdown_contents = []
for url in urls:
try:
content = markdown_tool._run(url=url)
markdown_contents.append(content)
except Exception as e:
print(f"Error processing {url}: {str(e)}")
continue
# Combine contents
combined_markdown = "\n\n---\n\n".join(markdown_contents)
```
## Notes
- Requires valid Serply API key
- Supports multiple proxy locations
- Returns markdown-formatted content
- Simplifies web content for LLM processing
- Thread-safe operations
- Efficient content conversion
- Handles API rate limiting automatically
- Preserves content structure in markdown
- Supports various webpage formats
- Makes web content more accessible to AI agents

View File

@@ -0,0 +1,158 @@
---
title: TXTSearchTool
description: A semantic search tool for text files using RAG capabilities
icon: magnifying-glass-document
---
## TXTSearchTool
The TXTSearchTool is a specialized Retrieval-Augmented Generation (RAG) tool that enables semantic search within text files. It inherits from the base RagTool class and provides both fixed and dynamic text file searching capabilities.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import TXTSearchTool
# Method 1: Dynamic file path
txt_search = TXTSearchTool()
# Method 2: Fixed file path
fixed_txt_search = TXTSearchTool(txt="path/to/fixed/document.txt")
# Create an agent with the tool
researcher = Agent(
role='Research Assistant',
goal='Search through text documents semantically',
backstory='Expert at finding relevant information in documents using semantic search.',
tools=[txt_search],
verbose=True
)
```
## Input Schema
The tool supports two input schemas depending on initialization:
### Dynamic File Path Schema
```python
class TXTSearchToolSchema(BaseModel):
search_query: str # The semantic search query
txt: str # Path to the text file to search
```
### Fixed File Path Schema
```python
class FixedTXTSearchToolSchema(BaseModel):
search_query: str # The semantic search query
```
## Function Signature
```python
def __init__(self, txt: Optional[str] = None, **kwargs):
"""
Initialize the TXT search tool.
Args:
txt (Optional[str]): Fixed path to a text file. If provided, the tool will only search this file.
**kwargs: Additional arguments passed to the parent RagTool
"""
def _run(self, search_query: str, **kwargs: Any) -> Any:
"""
Perform semantic search on the text file.
Args:
search_query (str): The semantic search query
**kwargs: Additional arguments (including 'txt' for dynamic file path)
Returns:
str: Relevant text passages based on semantic search
"""
```
## Best Practices
1. Choose initialization method based on use case:
- Use fixed file path when repeatedly searching the same document
- Use dynamic file path when searching different documents
2. Write clear, semantic search queries
3. Handle potential file access errors in agent prompts
4. Consider memory usage for large text files
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import TXTSearchTool
# Example 1: Fixed document search
documentation_search = TXTSearchTool(txt="api_documentation.txt")
# Example 2: Dynamic document search
flexible_search = TXTSearchTool()
# Create agents
doc_analyst = Agent(
role='Documentation Analyst',
goal='Find relevant API documentation sections',
backstory='Expert at analyzing technical documentation.',
tools=[documentation_search]
)
file_analyst = Agent(
role='File Analyst',
goal='Search through various text files',
backstory='Specialist in finding information across multiple documents.',
tools=[flexible_search]
)
# Define tasks
fixed_search_task = Task(
description="""Find all API endpoints related to user authentication
in the documentation.""",
agent=doc_analyst
)
# The agent will use:
# {
# "search_query": "user authentication API endpoints"
# }
dynamic_search_task = Task(
description="""Search through the logs.txt file for any database
connection errors.""",
agent=file_analyst
)
# The agent will use:
# {
# "search_query": "database connection errors",
# "txt": "logs.txt"
# }
# Create crew
crew = Crew(
agents=[doc_analyst, file_analyst],
tasks=[fixed_search_task, dynamic_search_task]
)
# Execute
result = crew.kickoff()
```
## Notes
- Inherits from RagTool for semantic search capabilities
- Supports both fixed and dynamic text file paths
- Uses embeddings for semantic search
- Optimized for text file analysis
- Thread-safe operations
- Automatically handles file loading and embedding

View File

@@ -0,0 +1,159 @@
---
title: YoutubeChannelSearchTool
description: A semantic search tool for YouTube channel content using RAG capabilities
icon: youtube
---
## YoutubeChannelSearchTool
The YoutubeChannelSearchTool is a specialized Retrieval-Augmented Generation (RAG) tool that enables semantic search within YouTube channel content. It inherits from the base RagTool class and provides both fixed and dynamic YouTube channel searching capabilities.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import YoutubeChannelSearchTool
# Method 1: Dynamic channel handle
youtube_search = YoutubeChannelSearchTool()
# Method 2: Fixed channel handle
fixed_channel_search = YoutubeChannelSearchTool(youtube_channel_handle="@example_channel")
# Create an agent with the tool
researcher = Agent(
role='Content Researcher',
goal='Search through YouTube channel content semantically',
backstory='Expert at finding relevant information in YouTube content.',
tools=[youtube_search],
verbose=True
)
```
## Input Schema
The tool supports two input schemas depending on initialization:
### Dynamic Channel Schema
```python
class YoutubeChannelSearchToolSchema(BaseModel):
search_query: str # The semantic search query
youtube_channel_handle: str # YouTube channel handle (with or without @)
```
### Fixed Channel Schema
```python
class FixedYoutubeChannelSearchToolSchema(BaseModel):
search_query: str # The semantic search query
```
## Function Signature
```python
def __init__(self, youtube_channel_handle: Optional[str] = None, **kwargs):
"""
Initialize the YouTube channel search tool.
Args:
youtube_channel_handle (Optional[str]): Fixed channel handle. If provided,
the tool will only search this channel.
**kwargs: Additional arguments passed to the parent RagTool
"""
def _run(self, search_query: str, **kwargs: Any) -> Any:
"""
Perform semantic search on the YouTube channel content.
Args:
search_query (str): The semantic search query
**kwargs: Additional arguments (including 'youtube_channel_handle' for dynamic mode)
Returns:
str: Relevant content from the YouTube channel based on semantic search
"""
```
## Best Practices
1. Choose initialization method based on use case:
- Use fixed channel handle when repeatedly searching the same channel
- Use dynamic handle when searching different channels
2. Write clear, semantic search queries
3. Channel handles can be provided with or without '@' prefix
4. Consider content availability and channel size
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import YoutubeChannelSearchTool
# Example 1: Fixed channel search
tech_channel_search = YoutubeChannelSearchTool(youtube_channel_handle="@TechChannel")
# Example 2: Dynamic channel search
flexible_search = YoutubeChannelSearchTool()
# Create agents
tech_analyst = Agent(
role='Tech Content Analyst',
goal='Find relevant tech tutorials and explanations',
backstory='Expert at analyzing technical YouTube content.',
tools=[tech_channel_search]
)
content_researcher = Agent(
role='Content Researcher',
goal='Search across multiple YouTube channels',
backstory='Specialist in finding information across various channels.',
tools=[flexible_search]
)
# Define tasks
fixed_search_task = Task(
description="""Find all tutorials related to machine learning
basics in the channel.""",
agent=tech_analyst
)
# The agent will use:
# {
# "search_query": "machine learning basics tutorial"
# }
dynamic_search_task = Task(
description="""Search through the @AIResearch channel for
content about neural networks.""",
agent=content_researcher
)
# The agent will use:
# {
# "search_query": "neural networks explanation",
# "youtube_channel_handle": "@AIResearch"
# }
# Create crew
crew = Crew(
agents=[tech_analyst, content_researcher],
tasks=[fixed_search_task, dynamic_search_task]
)
# Execute
result = crew.kickoff()
```
## Notes
- Inherits from RagTool for semantic search capabilities
- Supports both fixed and dynamic YouTube channel handles
- Automatically adds '@' prefix to channel handles if missing
- Uses embeddings for semantic search
- Thread-safe operations
- Automatically handles YouTube content loading and embedding

View File

@@ -0,0 +1,216 @@
---
title: YoutubeVideoSearchTool
description: A tool for semantic search within YouTube video content using RAG capabilities
icon: video
---
## YoutubeVideoSearchTool
The YoutubeVideoSearchTool enables semantic search capabilities for YouTube video content using Retrieval-Augmented Generation (RAG). It processes video content and allows searching through transcripts and metadata using natural language queries.
## Installation
```bash
pip install 'crewai[tools]'
```
## Usage Example
```python
from crewai import Agent
from crewai_tools import YoutubeVideoSearchTool
# Method 1: Initialize with specific video
video_tool = YoutubeVideoSearchTool(
youtube_video_url="https://www.youtube.com/watch?v=example"
)
# Method 2: Initialize without video (specify at runtime)
flexible_video_tool = YoutubeVideoSearchTool()
# Create an agent with the tool
researcher = Agent(
role='Video Researcher',
goal='Search and analyze video content',
backstory='Expert at finding relevant information in videos.',
tools=[video_tool],
verbose=True
)
```
## Input Schema
### Fixed Video Schema (when URL provided during initialization)
```python
class FixedYoutubeVideoSearchToolSchema(BaseModel):
search_query: str = Field(
description="Mandatory search query you want to use to search the Youtube Video content"
)
```
### Flexible Video Schema (when URL provided at runtime)
```python
class YoutubeVideoSearchToolSchema(FixedYoutubeVideoSearchToolSchema):
youtube_video_url: str = Field(
description="Mandatory youtube_video_url path you want to search"
)
```
## Function Signature
```python
def __init__(
self,
youtube_video_url: Optional[str] = None,
**kwargs
):
"""
Initialize the YouTube video search tool.
Args:
youtube_video_url (Optional[str]): URL of YouTube video (optional)
**kwargs: Additional arguments for RAG tool configuration
"""
def _run(
self,
search_query: str,
**kwargs: Any
) -> str:
"""
Execute semantic search on video content.
Args:
search_query (str): Query to search in the video
**kwargs: Additional arguments including youtube_video_url if not initialized
Returns:
str: Relevant content from the video matching the query
"""
```
## Best Practices
1. Video URL Management:
- Use complete YouTube URLs
- Verify video accessibility
- Handle region restrictions
2. Search Optimization:
- Use specific, focused queries
- Consider video context
- Test with sample queries first
3. Performance Considerations:
- Pre-initialize for repeated searches
- Handle long videos appropriately
- Monitor processing time
4. Error Handling:
- Verify video availability
- Handle unavailable videos
- Manage API limitations
## Integration Example
```python
from crewai import Agent, Task, Crew
from crewai_tools import YoutubeVideoSearchTool
# Initialize tool with specific video
video_tool = YoutubeVideoSearchTool(
youtube_video_url="https://www.youtube.com/watch?v=example"
)
# Create agent
researcher = Agent(
role='Video Researcher',
goal='Extract insights from video content',
backstory='Expert at analyzing video content.',
tools=[video_tool]
)
# Define task
research_task = Task(
description="""Find all mentions of machine learning
applications from the video content.""",
agent=researcher
)
# The tool will use:
# {
# "search_query": "machine learning applications"
# }
# Create crew
crew = Crew(
agents=[researcher],
tasks=[research_task]
)
# Execute
result = crew.kickoff()
```
## Advanced Usage
### Dynamic Video Selection
```python
# Initialize without video URL
flexible_tool = YoutubeVideoSearchTool()
# Search different videos
tech_results = flexible_tool.run(
search_query="quantum computing",
youtube_video_url="https://youtube.com/watch?v=tech123"
)
science_results = flexible_tool.run(
search_query="particle physics",
youtube_video_url="https://youtube.com/watch?v=science456"
)
```
### Multiple Video Analysis
```python
# Create tools for different videos
tech_tool = YoutubeVideoSearchTool(
youtube_video_url="https://youtube.com/watch?v=tech123"
)
science_tool = YoutubeVideoSearchTool(
youtube_video_url="https://youtube.com/watch?v=science456"
)
# Create agent with multiple tools
analyst = Agent(
role='Content Analyst',
goal='Cross-reference multiple videos',
tools=[tech_tool, science_tool]
)
```
### Error Handling Example
```python
try:
video_tool = YoutubeVideoSearchTool()
results = video_tool.run(
search_query="key concepts",
youtube_video_url="https://youtube.com/watch?v=example"
)
print(results)
except Exception as e:
print(f"Error processing video: {str(e)}")
```
## Notes
- Inherits from RagTool
- Uses embedchain for processing
- Supports semantic search
- Dynamic video specification
- Efficient content retrieval
- Thread-safe operations
- Maintains search context
- Handles video transcripts
- Processes video metadata
- Memory-efficient processing

View File

@@ -3,33 +3,44 @@ name = "crewai"
version = "0.86.0"
description = "Cutting-edge framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks."
readme = "README.md"
requires-python = ">=3.10,<=3.12"
requires-python = ">=3.10,<3.13"
authors = [
{ name = "Joao Moura", email = "joao@crewai.com" }
]
dependencies = [
# Core Dependencies
"pydantic>=2.4.2",
"openai>=1.13.3",
"litellm>=1.44.22",
"instructor>=1.3.3",
# Text Processing
"pdfplumber>=0.11.4",
"regex>=2024.9.11",
# Telemetry and Monitoring
"opentelemetry-api>=1.22.0",
"opentelemetry-sdk>=1.22.0",
"opentelemetry-exporter-otlp-proto-http>=1.22.0",
"instructor>=1.3.3",
"regex>=2024.9.11",
"crewai-tools>=0.17.0",
"click>=8.1.7",
# Data Handling
"chromadb>=0.5.23",
"openpyxl>=3.1.5",
"pyvis>=0.3.2",
# Authentication and Security
"auth0-python>=4.7.1",
"python-dotenv>=1.0.0",
# Configuration and Utils
"click>=8.1.7",
"appdirs>=1.4.4",
"jsonref>=1.1.0",
"json-repair>=0.25.2",
"auth0-python>=4.7.1",
"litellm>=1.44.22",
"pyvis>=0.3.2",
"uv>=0.4.25",
"tomli-w>=1.1.0",
"tomli>=2.0.2",
"chromadb>=0.5.18",
"pdfplumber>=0.11.4",
"openpyxl>=3.1.5",
"blinker>=1.9.0",
]
[project.urls]
@@ -38,7 +49,10 @@ Documentation = "https://docs.crewai.com"
Repository = "https://github.com/crewAIInc/crewAI"
[project.optional-dependencies]
tools = ["crewai-tools>=0.14.0"]
tools = ["crewai-tools>=0.17.0"]
embeddings = [
"tiktoken~=0.7.0"
]
agentops = ["agentops>=0.3.0"]
fastembed = ["fastembed>=0.4.1"]
pdfplumber = [
@@ -51,10 +65,13 @@ openpyxl = [
"openpyxl>=3.1.5",
]
mem0 = ["mem0ai>=0.1.29"]
docling = [
"docling>=2.12.0",
]
[tool.uv]
dev-dependencies = [
"ruff>=0.4.10",
"ruff>=0.8.2",
"mypy>=1.10.0",
"pre-commit>=3.6.0",
"mkdocs>=1.4.3",
@@ -64,7 +81,6 @@ dev-dependencies = [
"mkdocs-material-extensions>=1.3.1",
"pillow>=10.2.0",
"cairosvg>=2.7.1",
"crewai-tools>=0.14.0",
"pytest>=8.0.0",
"pytest-vcr>=1.0.2",
"python-dotenv>=1.0.0",

View File

@@ -17,33 +17,26 @@ from crewai.memory.contextual.contextual_memory import ContextualMemory
from crewai.task import Task
from crewai.tools import BaseTool
from crewai.tools.agent_tools.agent_tools import AgentTools
from crewai.tools.base_tool import Tool
from crewai.utilities import Converter, Prompts
from crewai.utilities.constants import TRAINED_AGENTS_DATA_FILE, TRAINING_DATA_FILE
from crewai.utilities.converter import generate_model_description
from crewai.utilities.token_counter_callback import TokenCalcHandler
from crewai.utilities.training_handler import CrewTrainingHandler
agentops = None
def mock_agent_ops_provider():
def track_agent(*args, **kwargs):
try:
import agentops # type: ignore # Name "agentops" is already defined
from agentops import track_agent # type: ignore
except ImportError:
def track_agent():
def noop(f):
return f
return noop
return track_agent
agentops = None
if os.environ.get("AGENTOPS_API_KEY"):
try:
from agentops import track_agent
except ImportError:
track_agent = mock_agent_ops_provider()
else:
track_agent = mock_agent_ops_provider()
@track_agent()
class Agent(BaseAgent):
@@ -122,6 +115,10 @@ class Agent(BaseAgent):
default=2,
description="Maximum number of retries for an agent to execute a task when an error occurs.",
)
multimodal: bool = Field(
default=False,
description="Whether the agent is multimodal.",
)
code_execution_mode: Literal["safe", "unsafe"] = Field(
default="safe",
description="Mode for code execution: 'safe' (using Docker) or 'unsafe' (direct execution).",
@@ -414,6 +411,10 @@ class Agent(BaseAgent):
tools = agent_tools.tools()
return tools
def get_multimodal_tools(self) -> List[Tool]:
from crewai.tools.agent_tools.add_image_tool import AddImageTool
return [AddImageTool()]
def get_code_execution_tools(self):
try:
from crewai_tools import CodeInterpreterTool

View File

@@ -143,10 +143,20 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
tool_result = self._execute_tool_and_check_finality(
formatted_answer
)
if self.step_callback:
self.step_callback(tool_result)
formatted_answer.text += f"\nObservation: {tool_result.result}"
# Directly append the result to the messages if the
# tool is "Add image to content" in case of multimodal
# agents
if formatted_answer.tool == self._i18n.tools("add_image")["name"]:
self.messages.append(tool_result.result)
continue
else:
if self.step_callback:
self.step_callback(tool_result)
formatted_answer.text += f"\nObservation: {tool_result.result}"
formatted_answer.result = tool_result.result
if tool_result.result_as_answer:
return AgentFinish(
@@ -413,7 +423,6 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
"""
while self.ask_for_human_input:
human_feedback = self._ask_human_input(formatted_answer.output)
print("Human feedback: ", human_feedback)
if self.crew and self.crew._train:
self._handle_crew_training_output(formatted_answer, human_feedback)

View File

@@ -1,5 +1,6 @@
import re
from typing import Any, Union
from json_repair import repair_json
from crewai.utilities import I18N

View File

@@ -5,9 +5,10 @@ from typing import Any, Dict
import requests
from rich.console import Console
from crewai.cli.tools.main import ToolCommand
from .constants import AUTH0_AUDIENCE, AUTH0_CLIENT_ID, AUTH0_DOMAIN
from .utils import TokenManager, validate_token
from crewai.cli.tools.main import ToolCommand
console = Console()
@@ -79,7 +80,9 @@ class AuthenticationCommand:
style="yellow",
)
console.print("\n[bold green]Welcome to CrewAI Enterprise![/bold green]\n")
console.print(
"\n[bold green]Welcome to CrewAI Enterprise![/bold green]\n"
)
return
if token_data["error"] not in ("authorization_pending", "slow_down"):

View File

@@ -1,10 +1,9 @@
from .utils import TokenManager
def get_auth_token() -> str:
"""Get the authentication token."""
access_token = TokenManager().get_token()
if not access_token:
raise Exception()
return access_token

View File

@@ -1,7 +1,7 @@
from importlib.metadata import version as get_version
from typing import Optional
import click
import pkg_resources
from crewai.cli.add_crew_to_flow import add_crew_to_flow
from crewai.cli.create_crew import create_crew
@@ -25,7 +25,7 @@ from .update_crew import update_crew
@click.group()
@click.version_option(pkg_resources.get_distribution("crewai").version)
@click.version_option(get_version("crewai"))
def crewai():
"""Top-level command group for crewai."""
@@ -52,16 +52,16 @@ def create(type, name, provider, skip_provider=False):
def version(tools):
"""Show the installed version of crewai."""
try:
crewai_version = pkg_resources.get_distribution("crewai").version
crewai_version = get_version("crewai")
except Exception:
crewai_version = "unknown version"
click.echo(f"crewai version: {crewai_version}")
if tools:
try:
tools_version = pkg_resources.get_distribution("crewai-tools").version
tools_version = get_version("crewai")
click.echo(f"crewai tools version: {tools_version}")
except pkg_resources.DistributionNotFound:
except Exception:
click.echo("crewai tools not installed")

View File

@@ -1,8 +1,9 @@
import requests
from requests.exceptions import JSONDecodeError
from rich.console import Console
from crewai.cli.plus_api import PlusAPI
from crewai.cli.authentication.token import get_auth_token
from crewai.cli.plus_api import PlusAPI
from crewai.telemetry.telemetry import Telemetry
console = Console()

View File

@@ -1,13 +1,19 @@
import json
from pathlib import Path
from pydantic import BaseModel, Field
from typing import Optional
from pydantic import BaseModel, Field
DEFAULT_CONFIG_PATH = Path.home() / ".config" / "crewai" / "settings.json"
class Settings(BaseModel):
tool_repository_username: Optional[str] = Field(None, description="Username for interacting with the Tool Repository")
tool_repository_password: Optional[str] = Field(None, description="Password for interacting with the Tool Repository")
tool_repository_username: Optional[str] = Field(
None, description="Username for interacting with the Tool Repository"
)
tool_repository_password: Optional[str] = Field(
None, description="Password for interacting with the Tool Repository"
)
config_path: Path = Field(default=DEFAULT_CONFIG_PATH, exclude=True)
def __init__(self, config_path: Path = DEFAULT_CONFIG_PATH, **data):

View File

@@ -1,9 +1,11 @@
from typing import Optional
import requests
from os import getenv
from crewai.cli.version import get_crewai_version
from typing import Optional
from urllib.parse import urljoin
import requests
from crewai.cli.version import get_crewai_version
class PlusAPI:
"""

View File

@@ -1,11 +1,12 @@
import subprocess
import click
from crewai.knowledge.storage.knowledge_storage import KnowledgeStorage
from crewai.memory.entity.entity_memory import EntityMemory
from crewai.memory.long_term.long_term_memory import LongTermMemory
from crewai.memory.short_term.short_term_memory import ShortTermMemory
from crewai.utilities.task_output_storage_handler import TaskOutputStorageHandler
from crewai.knowledge.storage.knowledge_storage import KnowledgeStorage
def reset_memories_command(

View File

@@ -4,7 +4,7 @@ Welcome to the {{crew_name}} Crew project, powered by [crewAI](https://crewai.co
## Installation
Ensure you have Python >=3.10 <=3.12 installed on your system. This project uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.
Ensure you have Python >=3.10 <3.13 installed on your system. This project uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.
First, if you haven't already, install uv:

View File

@@ -3,7 +3,7 @@ name = "{{folder_name}}"
version = "0.1.0"
description = "{{name}} using crewAI"
authors = [{ name = "Your Name", email = "you@example.com" }]
requires-python = ">=3.10,<=3.12"
requires-python = ">=3.10,<3.13"
dependencies = [
"crewai[tools]>=0.86.0,<1.0.0"
]
@@ -18,3 +18,6 @@ test = "{{folder_name}}.main:test"
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"
[tool.crewai]
type = "crew"

View File

@@ -10,7 +10,7 @@ class MyCustomToolInput(BaseModel):
class MyCustomTool(BaseTool):
name: str = "Name of my tool"
description: str = (
"Clear description for what this tool is useful for, you agent will need this information to use it."
"Clear description for what this tool is useful for, your agent will need this information to use it."
)
args_schema: Type[BaseModel] = MyCustomToolInput

View File

@@ -4,7 +4,7 @@ Welcome to the {{crew_name}} Crew project, powered by [crewAI](https://crewai.co
## Installation
Ensure you have Python >=3.10 <=3.12 installed on your system. This project uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.
Ensure you have Python >=3.10 <3.13 installed on your system. This project uses [UV](https://docs.astral.sh/uv/) for dependency management and package handling, offering a seamless setup and execution experience.
First, if you haven't already, install uv:

View File

@@ -5,7 +5,7 @@ from pydantic import BaseModel
from crewai.flow.flow import Flow, listen, start
from .crews.poem_crew.poem_crew import PoemCrew
from {{folder_name}}.crews.poem_crew.poem_crew import PoemCrew
class PoemState(BaseModel):

View File

@@ -3,7 +3,7 @@ name = "{{folder_name}}"
version = "0.1.0"
description = "{{name}} using crewAI"
authors = [{ name = "Your Name", email = "you@example.com" }]
requires-python = ">=3.10,<=3.12"
requires-python = ">=3.10,<3.13"
dependencies = [
"crewai[tools]>=0.86.0,<1.0.0",
]
@@ -15,3 +15,6 @@ plot = "{{folder_name}}.main:plot"
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"
[tool.crewai]
type = "flow"

View File

@@ -13,7 +13,7 @@ class MyCustomToolInput(BaseModel):
class MyCustomTool(BaseTool):
name: str = "Name of my tool"
description: str = (
"Clear description for what this tool is useful for, you agent will need this information to use it."
"Clear description for what this tool is useful for, your agent will need this information to use it."
)
args_schema: Type[BaseModel] = MyCustomToolInput

View File

@@ -1,17 +0,0 @@
[tool.poetry]
name = "{{folder_name}}"
version = "0.1.0"
description = "{{name}} using crewAI"
authors = ["Your Name <you@example.com>"]
[tool.poetry.dependencies]
python = ">=3.10,<=3.12"
crewai = { extras = ["tools"], version = ">=0.86.0,<1.0.0" }
asyncio = "*"
[tool.poetry.scripts]
{{folder_name}} = "{{folder_name}}.main:main"
[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"

View File

@@ -5,7 +5,7 @@ custom tools to power up your crews.
## Installing
Ensure you have Python >=3.10 <=3.12 installed on your system. This project
Ensure you have Python >=3.10 <3.13 installed on your system. This project
uses [UV](https://docs.astral.sh/uv/) for dependency management and package
handling, offering a seamless setup and execution experience.

View File

@@ -3,8 +3,10 @@ name = "{{folder_name}}"
version = "0.1.0"
description = "Power up your crews with {{folder_name}}"
readme = "README.md"
requires-python = ">=3.10,<=3.12"
requires-python = ">=3.10,<3.13"
dependencies = [
"crewai[tools]>=0.86.0"
]
[tool.crewai]
type = "tool"

View File

@@ -117,7 +117,7 @@ class ToolCommand(BaseCommand, PlusAPIMixin):
published_handle = publish_response.json()["handle"]
console.print(
f"Succesfully published {published_handle} ({project_version}).\nInstall it in other projects with crewai tool install {published_handle}",
f"Successfully published {published_handle} ({project_version}).\nInstall it in other projects with crewai tool install {published_handle}",
style="bold green",
)
@@ -138,7 +138,7 @@ class ToolCommand(BaseCommand, PlusAPIMixin):
self._add_package(get_response.json())
console.print(f"Succesfully installed {handle}", style="bold green")
console.print(f"Successfully installed {handle}", style="bold green")
def login(self):
login_response = self.plus_api_client.login_to_tool_repository()

View File

@@ -1,6 +1,6 @@
import importlib.metadata
def get_crewai_version() -> str:
"""Get the version number of CrewAI running the CLI"""
return importlib.metadata.version("crewai")

View File

@@ -1,6 +1,5 @@
import asyncio
import json
import os
import uuid
import warnings
from concurrent.futures import Future
@@ -36,6 +35,7 @@ from crewai.tasks.conditional_task import ConditionalTask
from crewai.tasks.task_output import TaskOutput
from crewai.telemetry import Telemetry
from crewai.tools.agent_tools.agent_tools import AgentTools
from crewai.tools.base_tool import Tool
from crewai.types.usage_metrics import UsageMetrics
from crewai.utilities import I18N, FileHandler, Logger, RPMController
from crewai.utilities.constants import TRAINING_DATA_FILE
@@ -49,12 +49,10 @@ from crewai.utilities.planning_handler import CrewPlanner
from crewai.utilities.task_output_storage_handler import TaskOutputStorageHandler
from crewai.utilities.training_handler import CrewTrainingHandler
agentops = None
if os.environ.get("AGENTOPS_API_KEY"):
try:
import agentops # type: ignore
except ImportError:
pass
try:
import agentops # type: ignore
except ImportError:
agentops = None
warnings.filterwarnings("ignore", category=SyntaxWarning, module="pysbd")
@@ -536,9 +534,6 @@ class Crew(BaseModel):
if not agent.function_calling_llm: # type: ignore # "BaseAgent" has no attribute "function_calling_llm"
agent.function_calling_llm = self.function_calling_llm # type: ignore # "BaseAgent" has no attribute "function_calling_llm"
if agent.allow_code_execution: # type: ignore # BaseAgent" has no attribute "allow_code_execution"
agent.tools += agent.get_code_execution_tools() # type: ignore # "BaseAgent" has no attribute "get_code_execution_tools"; maybe "get_delegation_tools"?
if not agent.step_callback: # type: ignore # "BaseAgent" has no attribute "step_callback"
agent.step_callback = self.step_callback # type: ignore # "BaseAgent" has no attribute "step_callback"
@@ -675,7 +670,6 @@ class Crew(BaseModel):
)
manager.tools = []
raise Exception("Manager agent should not have tools")
manager.tools = self.manager_agent.get_delegation_tools(self.agents)
else:
self.manager_llm = (
getattr(self.manager_llm, "model_name", None)
@@ -687,6 +681,7 @@ class Crew(BaseModel):
goal=i18n.retrieve("hierarchical_manager_agent", "goal"),
backstory=i18n.retrieve("hierarchical_manager_agent", "backstory"),
tools=AgentTools(agents=self.agents).tools(),
allow_delegation=True,
llm=self.manager_llm,
verbose=self.verbose,
)
@@ -729,7 +724,14 @@ class Crew(BaseModel):
f"No agent available for task: {task.description}. Ensure that either the task has an assigned agent or a manager agent is provided."
)
self._prepare_agent_tools(task)
# Determine which tools to use - task tools take precedence over agent tools
tools_for_task = task.tools or agent_to_use.tools or []
tools_for_task = self._prepare_tools(
agent_to_use,
task,
tools_for_task
)
self._log_task_start(task, agent_to_use.role)
if isinstance(task, ConditionalTask):
@@ -746,7 +748,7 @@ class Crew(BaseModel):
future = task.execute_async(
agent=agent_to_use,
context=context,
tools=agent_to_use.tools,
tools=tools_for_task,
)
futures.append((task, future, task_index))
else:
@@ -758,7 +760,7 @@ class Crew(BaseModel):
task_output = task.execute_sync(
agent=agent_to_use,
context=context,
tools=agent_to_use.tools,
tools=tools_for_task,
)
task_outputs = [task_output]
self._process_task_result(task, task_output)
@@ -795,45 +797,67 @@ class Crew(BaseModel):
return skipped_task_output
return None
def _prepare_agent_tools(self, task: Task):
if self.process == Process.hierarchical:
if self.manager_agent:
self._update_manager_tools(task)
else:
raise ValueError("Manager agent is required for hierarchical process.")
elif task.agent and task.agent.allow_delegation:
self._add_delegation_tools(task)
def _prepare_tools(self, agent: BaseAgent, task: Task, tools: List[Tool]) -> List[Tool]:
# Add delegation tools if agent allows delegation
if agent.allow_delegation:
if self.process == Process.hierarchical:
if self.manager_agent:
tools = self._update_manager_tools(task, tools)
else:
raise ValueError("Manager agent is required for hierarchical process.")
elif agent and agent.allow_delegation:
tools = self._add_delegation_tools(task, tools)
# Add code execution tools if agent allows code execution
if agent.allow_code_execution:
tools = self._add_code_execution_tools(agent, tools)
if agent and agent.multimodal:
tools = self._add_multimodal_tools(agent, tools)
return tools
def _get_agent_to_use(self, task: Task) -> Optional[BaseAgent]:
if self.process == Process.hierarchical:
return self.manager_agent
return task.agent
def _add_delegation_tools(self, task: Task):
def _merge_tools(self, existing_tools: List[Tool], new_tools: List[Tool]) -> List[Tool]:
"""Merge new tools into existing tools list, avoiding duplicates by tool name."""
if not new_tools:
return existing_tools
# Create mapping of tool names to new tools
new_tool_map = {tool.name: tool for tool in new_tools}
# Remove any existing tools that will be replaced
tools = [tool for tool in existing_tools if tool.name not in new_tool_map]
# Add all new tools
tools.extend(new_tools)
return tools
def _inject_delegation_tools(self, tools: List[Tool], task_agent: BaseAgent, agents: List[BaseAgent]):
delegation_tools = task_agent.get_delegation_tools(agents)
return self._merge_tools(tools, delegation_tools)
def _add_multimodal_tools(self, agent: BaseAgent, tools: List[Tool]):
multimodal_tools = agent.get_multimodal_tools()
return self._merge_tools(tools, multimodal_tools)
def _add_code_execution_tools(self, agent: BaseAgent, tools: List[Tool]):
code_tools = agent.get_code_execution_tools()
return self._merge_tools(tools, code_tools)
def _add_delegation_tools(self, task: Task, tools: List[Tool]):
agents_for_delegation = [agent for agent in self.agents if agent != task.agent]
if len(self.agents) > 1 and len(agents_for_delegation) > 0 and task.agent:
delegation_tools = task.agent.get_delegation_tools(agents_for_delegation)
# Add tools if they are not already in task.tools
for new_tool in delegation_tools:
# Find the index of the tool with the same name
existing_tool_index = next(
(
index
for index, tool in enumerate(task.tools or [])
if tool.name == new_tool.name
),
None,
)
if not task.tools:
task.tools = []
if existing_tool_index is not None:
# Replace the existing tool
task.tools[existing_tool_index] = new_tool
else:
# Add the new tool
task.tools.append(new_tool)
if not tools:
tools = []
tools = self._inject_delegation_tools(tools, task.agent, agents_for_delegation)
return tools
def _log_task_start(self, task: Task, role: str = "None"):
if self.output_log_file:
@@ -841,14 +865,13 @@ class Crew(BaseModel):
task_name=task.name, task=task.description, agent=role, status="started"
)
def _update_manager_tools(self, task: Task):
def _update_manager_tools(self, task: Task, tools: List[Tool]):
if self.manager_agent:
if task.agent:
self.manager_agent.tools = task.agent.get_delegation_tools([task.agent])
tools = self._inject_delegation_tools(tools, task.agent, [task.agent])
else:
self.manager_agent.tools = self.manager_agent.get_delegation_tools(
self.agents
)
tools = self._inject_delegation_tools(tools, self.manager_agent, self.agents)
return tools
def _get_context(self, task: Task, task_outputs: List[TaskOutput]):
context = (
@@ -1032,6 +1055,7 @@ class Crew(BaseModel):
agentops.end_session(
end_state="Success",
end_state_reason="Finished Execution",
is_auto_end=True,
)
self._telemetry.end_crew(self, final_string_output)

View File

@@ -14,8 +14,15 @@ from typing import (
cast,
)
from blinker import Signal
from pydantic import BaseModel, ValidationError
from crewai.flow.flow_events import (
FlowFinishedEvent,
FlowStartedEvent,
MethodExecutionFinishedEvent,
MethodExecutionStartedEvent,
)
from crewai.flow.flow_visualizer import plot_flow
from crewai.flow.utils import get_possible_return_constants
from crewai.telemetry import Telemetry
@@ -23,7 +30,47 @@ from crewai.telemetry import Telemetry
T = TypeVar("T", bound=Union[BaseModel, Dict[str, Any]])
def start(condition=None):
def start(condition: Optional[Union[str, dict, Callable]] = None) -> Callable:
"""
Marks a method as a flow's starting point.
This decorator designates a method as an entry point for the flow execution.
It can optionally specify conditions that trigger the start based on other
method executions.
Parameters
----------
condition : Optional[Union[str, dict, Callable]], optional
Defines when the start method should execute. Can be:
- str: Name of a method that triggers this start
- dict: Contains "type" ("AND"/"OR") and "methods" (list of triggers)
- Callable: A method reference that triggers this start
Default is None, meaning unconditional start.
Returns
-------
Callable
A decorator function that marks the method as a flow start point.
Raises
------
ValueError
If the condition format is invalid.
Examples
--------
>>> @start() # Unconditional start
>>> def begin_flow(self):
... pass
>>> @start("method_name") # Start after specific method
>>> def conditional_start(self):
... pass
>>> @start(and_("method1", "method2")) # Start after multiple methods
>>> def complex_start(self):
... pass
"""
def decorator(func):
func.__is_start_method__ = True
if condition is not None:
@@ -48,8 +95,42 @@ def start(condition=None):
return decorator
def listen(condition: Union[str, dict, Callable]) -> Callable:
"""
Creates a listener that executes when specified conditions are met.
def listen(condition):
This decorator sets up a method to execute in response to other method
executions in the flow. It supports both simple and complex triggering
conditions.
Parameters
----------
condition : Union[str, dict, Callable]
Specifies when the listener should execute. Can be:
- str: Name of a method that triggers this listener
- dict: Contains "type" ("AND"/"OR") and "methods" (list of triggers)
- Callable: A method reference that triggers this listener
Returns
-------
Callable
A decorator function that sets up the method as a listener.
Raises
------
ValueError
If the condition format is invalid.
Examples
--------
>>> @listen("process_data") # Listen to single method
>>> def handle_processed_data(self):
... pass
>>> @listen(or_("success", "failure")) # Listen to multiple methods
>>> def handle_completion(self):
... pass
"""
def decorator(func):
if isinstance(condition, str):
func.__trigger_methods__ = [condition]
@@ -73,16 +154,103 @@ def listen(condition):
return decorator
def router(method):
def router(condition: Union[str, dict, Callable]) -> Callable:
"""
Creates a routing method that directs flow execution based on conditions.
This decorator marks a method as a router, which can dynamically determine
the next steps in the flow based on its return value. Routers are triggered
by specified conditions and can return constants that determine which path
the flow should take.
Parameters
----------
condition : Union[str, dict, Callable]
Specifies when the router should execute. Can be:
- str: Name of a method that triggers this router
- dict: Contains "type" ("AND"/"OR") and "methods" (list of triggers)
- Callable: A method reference that triggers this router
Returns
-------
Callable
A decorator function that sets up the method as a router.
Raises
------
ValueError
If the condition format is invalid.
Examples
--------
>>> @router("check_status")
>>> def route_based_on_status(self):
... if self.state.status == "success":
... return SUCCESS
... return FAILURE
>>> @router(and_("validate", "process"))
>>> def complex_routing(self):
... if all([self.state.valid, self.state.processed]):
... return CONTINUE
... return STOP
"""
def decorator(func):
func.__is_router__ = True
func.__router_for__ = method.__name__
if isinstance(condition, str):
func.__trigger_methods__ = [condition]
func.__condition_type__ = "OR"
elif (
isinstance(condition, dict)
and "type" in condition
and "methods" in condition
):
func.__trigger_methods__ = condition["methods"]
func.__condition_type__ = condition["type"]
elif callable(condition) and hasattr(condition, "__name__"):
func.__trigger_methods__ = [condition.__name__]
func.__condition_type__ = "OR"
else:
raise ValueError(
"Condition must be a method, string, or a result of or_() or and_()"
)
return func
return decorator
def or_(*conditions: Union[str, dict, Callable]) -> dict:
"""
Combines multiple conditions with OR logic for flow control.
def or_(*conditions):
Creates a condition that is satisfied when any of the specified conditions
are met. This is used with @start, @listen, or @router decorators to create
complex triggering conditions.
Parameters
----------
*conditions : Union[str, dict, Callable]
Variable number of conditions that can be:
- str: Method names
- dict: Existing condition dictionaries
- Callable: Method references
Returns
-------
dict
A condition dictionary with format:
{"type": "OR", "methods": list_of_method_names}
Raises
------
ValueError
If any condition is invalid.
Examples
--------
>>> @listen(or_("success", "timeout"))
>>> def handle_completion(self):
... pass
"""
methods = []
for condition in conditions:
if isinstance(condition, dict) and "methods" in condition:
@@ -96,7 +264,39 @@ def or_(*conditions):
return {"type": "OR", "methods": methods}
def and_(*conditions):
def and_(*conditions: Union[str, dict, Callable]) -> dict:
"""
Combines multiple conditions with AND logic for flow control.
Creates a condition that is satisfied only when all specified conditions
are met. This is used with @start, @listen, or @router decorators to create
complex triggering conditions.
Parameters
----------
*conditions : Union[str, dict, Callable]
Variable number of conditions that can be:
- str: Method names
- dict: Existing condition dictionaries
- Callable: Method references
Returns
-------
dict
A condition dictionary with format:
{"type": "AND", "methods": list_of_method_names}
Raises
------
ValueError
If any condition is invalid.
Examples
--------
>>> @listen(and_("validated", "processed"))
>>> def handle_complete_data(self):
... pass
"""
methods = []
for condition in conditions:
if isinstance(condition, dict) and "methods" in condition:
@@ -116,8 +316,8 @@ class FlowMeta(type):
start_methods = []
listeners = {}
routers = {}
router_paths = {}
routers = set()
for attr_name, attr_value in dct.items():
if hasattr(attr_value, "__is_start_method__"):
@@ -130,18 +330,11 @@ class FlowMeta(type):
methods = attr_value.__trigger_methods__
condition_type = getattr(attr_value, "__condition_type__", "OR")
listeners[attr_name] = (condition_type, methods)
elif hasattr(attr_value, "__is_router__"):
routers[attr_value.__router_for__] = attr_name
possible_returns = get_possible_return_constants(attr_value)
if possible_returns:
router_paths[attr_name] = possible_returns
# Register router as a listener to its triggering method
trigger_method_name = attr_value.__router_for__
methods = [trigger_method_name]
condition_type = "OR"
listeners[attr_name] = (condition_type, methods)
if hasattr(attr_value, "__is_router__") and attr_value.__is_router__:
routers.add(attr_name)
possible_returns = get_possible_return_constants(attr_value)
if possible_returns:
router_paths[attr_name] = possible_returns
setattr(cls, "_start_methods", start_methods)
setattr(cls, "_listeners", listeners)
@@ -156,9 +349,10 @@ class Flow(Generic[T], metaclass=FlowMeta):
_start_methods: List[str] = []
_listeners: Dict[str, tuple[str, List[str]]] = {}
_routers: Dict[str, str] = {}
_routers: Set[str] = set()
_router_paths: Dict[str, List[str]] = {}
initial_state: Union[Type[T], T, None] = None
event_emitter = Signal("event_emitter")
def __class_getitem__(cls: Type["Flow"], item: Type[T]) -> Type["Flow"]:
class _FlowGeneric(cls): # type: ignore
@@ -202,20 +396,10 @@ class Flow(Generic[T], metaclass=FlowMeta):
return self._method_outputs
def _initialize_state(self, inputs: Dict[str, Any]) -> None:
"""
Initializes or updates the state with the provided inputs.
Args:
inputs: Dictionary of inputs to initialize or update the state.
Raises:
ValueError: If inputs do not match the structured state model.
TypeError: If state is neither a BaseModel instance nor a dictionary.
"""
if isinstance(self._state, BaseModel):
# Structured state management
# Structured state
try:
# Define a function to create the dynamic class
def create_model_with_extra_forbid(
base_model: Type[BaseModel],
) -> Type[BaseModel]:
@@ -225,50 +409,33 @@ class Flow(Generic[T], metaclass=FlowMeta):
return ModelWithExtraForbid
# Create the dynamic class
ModelWithExtraForbid = create_model_with_extra_forbid(
self._state.__class__
)
# Create a new instance using the combined state and inputs
self._state = cast(
T, ModelWithExtraForbid(**{**self._state.model_dump(), **inputs})
)
except ValidationError as e:
raise ValueError(f"Invalid inputs for structured state: {e}") from e
elif isinstance(self._state, dict):
# Unstructured state management
self._state.update(inputs)
else:
raise TypeError("State must be a BaseModel instance or a dictionary.")
def kickoff(self, inputs: Optional[Dict[str, Any]] = None) -> Any:
"""
Starts the execution of the flow synchronously.
self.event_emitter.send(
self,
event=FlowStartedEvent(
type="flow_started",
flow_name=self.__class__.__name__,
),
)
Args:
inputs: Optional dictionary of inputs to initialize or update the state.
Returns:
The final output from the flow execution.
"""
if inputs is not None:
self._initialize_state(inputs)
return asyncio.run(self.kickoff_async())
async def kickoff_async(self, inputs: Optional[Dict[str, Any]] = None) -> Any:
"""
Starts the execution of the flow asynchronously.
Args:
inputs: Optional dictionary of inputs to initialize or update the state.
Returns:
The final output from the flow execution.
"""
if inputs is not None:
self._initialize_state(inputs)
if not self._start_methods:
raise ValueError("No start method defined")
@@ -276,22 +443,42 @@ class Flow(Generic[T], metaclass=FlowMeta):
self.__class__.__name__, list(self._methods.keys())
)
# Create tasks for all start methods
tasks = [
self._execute_start_method(start_method)
for start_method in self._start_methods
]
# Run all start methods concurrently
await asyncio.gather(*tasks)
# Return the final output (from the last executed method)
if self._method_outputs:
return self._method_outputs[-1]
else:
return None # Or raise an exception if no methods were executed
final_output = self._method_outputs[-1] if self._method_outputs else None
self.event_emitter.send(
self,
event=FlowFinishedEvent(
type="flow_finished",
flow_name=self.__class__.__name__,
result=final_output,
),
)
return final_output
async def _execute_start_method(self, start_method_name: str) -> None:
"""
Executes a flow's start method and its triggered listeners.
This internal method handles the execution of methods marked with @start
decorator and manages the subsequent chain of listener executions.
Parameters
----------
start_method_name : str
The name of the start method to execute.
Notes
-----
- Executes the start method and captures its result
- Triggers execution of any listeners waiting on this start method
- Part of the flow's initialization sequence
"""
result = await self._execute_method(
start_method_name, self._methods[start_method_name]
)
@@ -305,70 +492,181 @@ class Flow(Generic[T], metaclass=FlowMeta):
if asyncio.iscoroutinefunction(method)
else method(*args, **kwargs)
)
self._method_outputs.append(result) # Store the output
# Track method execution counts
self._method_outputs.append(result)
self._method_execution_counts[method_name] = (
self._method_execution_counts.get(method_name, 0) + 1
)
return result
async def _execute_listeners(self, trigger_method: str, result: Any) -> None:
listener_tasks = []
"""
Executes all listeners and routers triggered by a method completion.
if trigger_method in self._routers:
router_method = self._methods[self._routers[trigger_method]]
path = await self._execute_method(
self._routers[trigger_method], router_method
This internal method manages the execution flow by:
1. First executing all triggered routers sequentially
2. Then executing all triggered listeners in parallel
Parameters
----------
trigger_method : str
The name of the method that triggered these listeners.
result : Any
The result from the triggering method, passed to listeners
that accept parameters.
Notes
-----
- Routers are executed sequentially to maintain flow control
- Each router's result becomes the new trigger_method
- Normal listeners are executed in parallel for efficiency
- Listeners can receive the trigger method's result as a parameter
"""
# First, handle routers repeatedly until no router triggers anymore
while True:
routers_triggered = self._find_triggered_methods(
trigger_method, router_only=True
)
trigger_method = path
if not routers_triggered:
break
for router_name in routers_triggered:
await self._execute_single_listener(router_name, result)
# After executing router, the router's result is the path
# The last router executed sets the trigger_method
# The router result is the last element in self._method_outputs
trigger_method = self._method_outputs[-1]
# Now that no more routers are triggered by current trigger_method,
# execute normal listeners
listeners_triggered = self._find_triggered_methods(
trigger_method, router_only=False
)
if listeners_triggered:
tasks = [
self._execute_single_listener(listener_name, result)
for listener_name in listeners_triggered
]
await asyncio.gather(*tasks)
def _find_triggered_methods(
self, trigger_method: str, router_only: bool
) -> List[str]:
"""
Finds all methods that should be triggered based on conditions.
This internal method evaluates both OR and AND conditions to determine
which methods should be executed next in the flow.
Parameters
----------
trigger_method : str
The name of the method that just completed execution.
router_only : bool
If True, only consider router methods.
If False, only consider non-router methods.
Returns
-------
List[str]
Names of methods that should be triggered.
Notes
-----
- Handles both OR and AND conditions:
* OR: Triggers if any condition is met
* AND: Triggers only when all conditions are met
- Maintains state for AND conditions using _pending_and_listeners
- Separates router and normal listener evaluation
"""
triggered = []
for listener_name, (condition_type, methods) in self._listeners.items():
is_router = listener_name in self._routers
if router_only != is_router:
continue
if condition_type == "OR":
# If the trigger_method matches any in methods, run this
if trigger_method in methods:
# Schedule the listener without preventing re-execution
listener_tasks.append(
self._execute_single_listener(listener_name, result)
)
triggered.append(listener_name)
elif condition_type == "AND":
# Initialize pending methods for this listener if not already done
if listener_name not in self._pending_and_listeners:
self._pending_and_listeners[listener_name] = set(methods)
# Remove the trigger method from pending methods
self._pending_and_listeners[listener_name].discard(trigger_method)
if trigger_method in self._pending_and_listeners[listener_name]:
self._pending_and_listeners[listener_name].discard(trigger_method)
if not self._pending_and_listeners[listener_name]:
# All required methods have been executed
listener_tasks.append(
self._execute_single_listener(listener_name, result)
)
triggered.append(listener_name)
# Reset pending methods for this listener
self._pending_and_listeners.pop(listener_name, None)
# Run all listener tasks concurrently and wait for them to complete
if listener_tasks:
await asyncio.gather(*listener_tasks)
return triggered
async def _execute_single_listener(self, listener_name: str, result: Any) -> None:
"""
Executes a single listener method with proper event handling.
This internal method manages the execution of an individual listener,
including parameter inspection, event emission, and error handling.
Parameters
----------
listener_name : str
The name of the listener method to execute.
result : Any
The result from the triggering method, which may be passed
to the listener if it accepts parameters.
Notes
-----
- Inspects method signature to determine if it accepts the trigger result
- Emits events for method execution start and finish
- Handles errors gracefully with detailed logging
- Recursively triggers listeners of this listener
- Supports both parameterized and parameter-less listeners
Error Handling
-------------
Catches and logs any exceptions during execution, preventing
individual listener failures from breaking the entire flow.
"""
try:
method = self._methods[listener_name]
self.event_emitter.send(
self,
event=MethodExecutionStartedEvent(
type="method_execution_started",
method_name=listener_name,
flow_name=self.__class__.__name__,
),
)
sig = inspect.signature(method)
params = list(sig.parameters.values())
# Exclude 'self' parameter
method_params = [p for p in params if p.name != "self"]
if method_params:
# If listener expects parameters, pass the result
listener_result = await self._execute_method(
listener_name, method, result
)
else:
# If listener does not expect parameters, call without arguments
listener_result = await self._execute_method(listener_name, method)
# Execute listeners of this listener
self.event_emitter.send(
self,
event=MethodExecutionFinishedEvent(
type="method_execution_finished",
method_name=listener_name,
flow_name=self.__class__.__name__,
),
)
# Execute listeners (and possibly routers) of this listener
await self._execute_listeners(listener_name, listener_result)
except Exception as e:
print(
f"[Flow._execute_single_listener] Error in method {listener_name}: {e}"
@@ -381,5 +679,4 @@ class Flow(Generic[T], metaclass=FlowMeta):
self._telemetry.flow_plotting_span(
self.__class__.__name__, list(self._methods.keys())
)
plot_flow(self, filename)

View File

@@ -0,0 +1,33 @@
from dataclasses import dataclass, field
from datetime import datetime
from typing import Any, Optional
@dataclass
class Event:
type: str
flow_name: str
timestamp: datetime = field(init=False)
def __post_init__(self):
self.timestamp = datetime.now()
@dataclass
class FlowStartedEvent(Event):
pass
@dataclass
class MethodExecutionStartedEvent(Event):
method_name: str
@dataclass
class MethodExecutionFinishedEvent(Event):
method_name: str
@dataclass
class FlowFinishedEvent(Event):
result: Optional[Any] = None

View File

@@ -1,12 +1,14 @@
# flow_visualizer.py
import os
from pathlib import Path
from pyvis.network import Network
from crewai.flow.config import COLORS, NODE_STYLES
from crewai.flow.html_template_handler import HTMLTemplateHandler
from crewai.flow.legend_generator import generate_legend_items_html, get_legend_items
from crewai.flow.path_utils import safe_path_join, validate_path_exists
from crewai.flow.utils import calculate_node_levels
from crewai.flow.visualization_utils import (
add_edges,
@@ -16,89 +18,209 @@ from crewai.flow.visualization_utils import (
class FlowPlot:
"""Handles the creation and rendering of flow visualization diagrams."""
def __init__(self, flow):
"""
Initialize FlowPlot with a flow object.
Parameters
----------
flow : Flow
A Flow instance to visualize.
Raises
------
ValueError
If flow object is invalid or missing required attributes.
"""
if not hasattr(flow, '_methods'):
raise ValueError("Invalid flow object: missing '_methods' attribute")
if not hasattr(flow, '_listeners'):
raise ValueError("Invalid flow object: missing '_listeners' attribute")
if not hasattr(flow, '_start_methods'):
raise ValueError("Invalid flow object: missing '_start_methods' attribute")
self.flow = flow
self.colors = COLORS
self.node_styles = NODE_STYLES
def plot(self, filename):
net = Network(
directed=True,
height="750px",
width="100%",
bgcolor=self.colors["bg"],
layout=None,
)
# Set options to disable physics
net.set_options(
"""
var options = {
"nodes": {
"font": {
"multi": "html"
}
},
"physics": {
"enabled": false
}
}
"""
)
Generate and save an HTML visualization of the flow.
# Calculate levels for nodes
node_levels = calculate_node_levels(self.flow)
Parameters
----------
filename : str
Name of the output file (without extension).
# Compute positions
node_positions = compute_positions(self.flow, node_levels)
Raises
------
ValueError
If filename is invalid or network generation fails.
IOError
If file operations fail or visualization cannot be generated.
RuntimeError
If network visualization generation fails.
"""
if not filename or not isinstance(filename, str):
raise ValueError("Filename must be a non-empty string")
try:
# Initialize network
net = Network(
directed=True,
height="750px",
width="100%",
bgcolor=self.colors["bg"],
layout=None,
)
# Add nodes to the network
add_nodes_to_network(net, self.flow, node_positions, self.node_styles)
# Set options to disable physics
net.set_options(
"""
var options = {
"nodes": {
"font": {
"multi": "html"
}
},
"physics": {
"enabled": false
}
}
"""
)
# Add edges to the network
add_edges(net, self.flow, node_positions, self.colors)
# Calculate levels for nodes
try:
node_levels = calculate_node_levels(self.flow)
except Exception as e:
raise ValueError(f"Failed to calculate node levels: {str(e)}")
network_html = net.generate_html()
final_html_content = self._generate_final_html(network_html)
# Compute positions
try:
node_positions = compute_positions(self.flow, node_levels)
except Exception as e:
raise ValueError(f"Failed to compute node positions: {str(e)}")
# Save the final HTML content to the file
with open(f"{filename}.html", "w", encoding="utf-8") as f:
f.write(final_html_content)
print(f"Plot saved as {filename}.html")
# Add nodes to the network
try:
add_nodes_to_network(net, self.flow, node_positions, self.node_styles)
except Exception as e:
raise RuntimeError(f"Failed to add nodes to network: {str(e)}")
self._cleanup_pyvis_lib()
# Add edges to the network
try:
add_edges(net, self.flow, node_positions, self.colors)
except Exception as e:
raise RuntimeError(f"Failed to add edges to network: {str(e)}")
# Generate HTML
try:
network_html = net.generate_html()
final_html_content = self._generate_final_html(network_html)
except Exception as e:
raise RuntimeError(f"Failed to generate network visualization: {str(e)}")
# Save the final HTML content to the file
try:
with open(f"{filename}.html", "w", encoding="utf-8") as f:
f.write(final_html_content)
print(f"Plot saved as {filename}.html")
except IOError as e:
raise IOError(f"Failed to save flow visualization to {filename}.html: {str(e)}")
except (ValueError, RuntimeError, IOError) as e:
raise e
except Exception as e:
raise RuntimeError(f"Unexpected error during flow visualization: {str(e)}")
finally:
self._cleanup_pyvis_lib()
def _generate_final_html(self, network_html):
# Extract just the body content from the generated HTML
current_dir = os.path.dirname(__file__)
template_path = os.path.join(
current_dir, "assets", "crewai_flow_visual_template.html"
)
logo_path = os.path.join(current_dir, "assets", "crewai_logo.svg")
"""
Generate the final HTML content with network visualization and legend.
html_handler = HTMLTemplateHandler(template_path, logo_path)
network_body = html_handler.extract_body_content(network_html)
Parameters
----------
network_html : str
HTML content generated by pyvis Network.
# Generate the legend items HTML
legend_items = get_legend_items(self.colors)
legend_items_html = generate_legend_items_html(legend_items)
final_html_content = html_handler.generate_final_html(
network_body, legend_items_html
)
return final_html_content
Returns
-------
str
Complete HTML content with styling and legend.
Raises
------
IOError
If template or logo files cannot be accessed.
ValueError
If network_html is invalid.
"""
if not network_html:
raise ValueError("Invalid network HTML content")
try:
# Extract just the body content from the generated HTML
current_dir = os.path.dirname(__file__)
template_path = safe_path_join("assets", "crewai_flow_visual_template.html", root=current_dir)
logo_path = safe_path_join("assets", "crewai_logo.svg", root=current_dir)
if not os.path.exists(template_path):
raise IOError(f"Template file not found: {template_path}")
if not os.path.exists(logo_path):
raise IOError(f"Logo file not found: {logo_path}")
html_handler = HTMLTemplateHandler(template_path, logo_path)
network_body = html_handler.extract_body_content(network_html)
# Generate the legend items HTML
legend_items = get_legend_items(self.colors)
legend_items_html = generate_legend_items_html(legend_items)
final_html_content = html_handler.generate_final_html(
network_body, legend_items_html
)
return final_html_content
except Exception as e:
raise IOError(f"Failed to generate visualization HTML: {str(e)}")
def _cleanup_pyvis_lib(self):
# Clean up the generated lib folder
lib_folder = os.path.join(os.getcwd(), "lib")
"""
Clean up the generated lib folder from pyvis.
This method safely removes the temporary lib directory created by pyvis
during network visualization generation.
"""
try:
lib_folder = safe_path_join("lib", root=os.getcwd())
if os.path.exists(lib_folder) and os.path.isdir(lib_folder):
import shutil
shutil.rmtree(lib_folder)
except ValueError as e:
print(f"Error validating lib folder path: {e}")
except Exception as e:
print(f"Error cleaning up {lib_folder}: {e}")
print(f"Error cleaning up lib folder: {e}")
def plot_flow(flow, filename="flow_plot"):
"""
Convenience function to create and save a flow visualization.
Parameters
----------
flow : Flow
Flow instance to visualize.
filename : str, optional
Output filename without extension, by default "flow_plot".
Raises
------
ValueError
If flow object or filename is invalid.
IOError
If file operations fail.
"""
visualizer = FlowPlot(flow)
visualizer.plot(filename)

View File

@@ -1,26 +1,53 @@
import base64
import re
from pathlib import Path
from crewai.flow.path_utils import safe_path_join, validate_path_exists
class HTMLTemplateHandler:
"""Handles HTML template processing and generation for flow visualization diagrams."""
def __init__(self, template_path, logo_path):
self.template_path = template_path
self.logo_path = logo_path
"""
Initialize HTMLTemplateHandler with validated template and logo paths.
Parameters
----------
template_path : str
Path to the HTML template file.
logo_path : str
Path to the logo image file.
Raises
------
ValueError
If template or logo paths are invalid or files don't exist.
"""
try:
self.template_path = validate_path_exists(template_path, "file")
self.logo_path = validate_path_exists(logo_path, "file")
except ValueError as e:
raise ValueError(f"Invalid template or logo path: {e}")
def read_template(self):
"""Read and return the HTML template file contents."""
with open(self.template_path, "r", encoding="utf-8") as f:
return f.read()
def encode_logo(self):
"""Convert the logo SVG file to base64 encoded string."""
with open(self.logo_path, "rb") as logo_file:
logo_svg_data = logo_file.read()
return base64.b64encode(logo_svg_data).decode("utf-8")
def extract_body_content(self, html):
"""Extract and return content between body tags from HTML string."""
match = re.search("<body.*?>(.*?)</body>", html, re.DOTALL)
return match.group(1) if match else ""
def generate_legend_items_html(self, legend_items):
"""Generate HTML markup for the legend items."""
legend_items_html = ""
for item in legend_items:
if "border" in item:
@@ -48,6 +75,7 @@ class HTMLTemplateHandler:
return legend_items_html
def generate_final_html(self, network_body, legend_items_html, title="Flow Plot"):
"""Combine all components into final HTML document with network visualization."""
html_template = self.read_template()
logo_svg_base64 = self.encode_logo()

View File

@@ -1,3 +1,4 @@
def get_legend_items(colors):
return [
{"label": "Start Method", "color": colors["start"]},

View File

@@ -0,0 +1,135 @@
"""
Path utilities for secure file operations in CrewAI flow module.
This module provides utilities for secure path handling to prevent directory
traversal attacks and ensure paths remain within allowed boundaries.
"""
import os
from pathlib import Path
from typing import List, Union
def safe_path_join(*parts: str, root: Union[str, Path, None] = None) -> str:
"""
Safely join path components and ensure the result is within allowed boundaries.
Parameters
----------
*parts : str
Variable number of path components to join.
root : Union[str, Path, None], optional
Root directory to use as base. If None, uses current working directory.
Returns
-------
str
String representation of the resolved path.
Raises
------
ValueError
If the resulting path would be outside the root directory
or if any path component is invalid.
"""
if not parts:
raise ValueError("No path components provided")
try:
# Convert all parts to strings and clean them
clean_parts = [str(part).strip() for part in parts if part]
if not clean_parts:
raise ValueError("No valid path components provided")
# Establish root directory
root_path = Path(root).resolve() if root else Path.cwd()
# Join and resolve the full path
full_path = Path(root_path, *clean_parts).resolve()
# Check if the resolved path is within root
if not str(full_path).startswith(str(root_path)):
raise ValueError(
f"Invalid path: Potential directory traversal. Path must be within {root_path}"
)
return str(full_path)
except Exception as e:
if isinstance(e, ValueError):
raise
raise ValueError(f"Invalid path components: {str(e)}")
def validate_path_exists(path: Union[str, Path], file_type: str = "file") -> str:
"""
Validate that a path exists and is of the expected type.
Parameters
----------
path : Union[str, Path]
Path to validate.
file_type : str, optional
Expected type ('file' or 'directory'), by default 'file'.
Returns
-------
str
Validated path as string.
Raises
------
ValueError
If path doesn't exist or is not of expected type.
"""
try:
path_obj = Path(path).resolve()
if not path_obj.exists():
raise ValueError(f"Path does not exist: {path}")
if file_type == "file" and not path_obj.is_file():
raise ValueError(f"Path is not a file: {path}")
elif file_type == "directory" and not path_obj.is_dir():
raise ValueError(f"Path is not a directory: {path}")
return str(path_obj)
except Exception as e:
if isinstance(e, ValueError):
raise
raise ValueError(f"Invalid path: {str(e)}")
def list_files(directory: Union[str, Path], pattern: str = "*") -> List[str]:
"""
Safely list files in a directory matching a pattern.
Parameters
----------
directory : Union[str, Path]
Directory to search in.
pattern : str, optional
Glob pattern to match files against, by default "*".
Returns
-------
List[str]
List of matching file paths.
Raises
------
ValueError
If directory is invalid or inaccessible.
"""
try:
dir_path = Path(directory).resolve()
if not dir_path.is_dir():
raise ValueError(f"Not a directory: {directory}")
return [str(p) for p in dir_path.glob(pattern) if p.is_file()]
except Exception as e:
if isinstance(e, ValueError):
raise
raise ValueError(f"Error listing files: {str(e)}")

View File

@@ -1,9 +1,25 @@
"""
Utility functions for flow visualization and dependency analysis.
This module provides core functionality for analyzing and manipulating flow structures,
including node level calculation, ancestor tracking, and return value analysis.
Functions in this module are primarily used by the visualization system to create
accurate and informative flow diagrams.
Example
-------
>>> flow = Flow()
>>> node_levels = calculate_node_levels(flow)
>>> ancestors = build_ancestor_dict(flow)
"""
import ast
import inspect
import textwrap
from typing import Any, Dict, List, Optional, Set, Union
def get_possible_return_constants(function):
def get_possible_return_constants(function: Any) -> Optional[List[str]]:
try:
source = inspect.getsource(function)
except OSError:
@@ -31,23 +47,80 @@ def get_possible_return_constants(function):
print(f"Source code:\n{source}")
return None
return_values = []
return_values = set()
dict_definitions = {}
class DictionaryAssignmentVisitor(ast.NodeVisitor):
def visit_Assign(self, node):
# Check if this assignment is assigning a dictionary literal to a variable
if isinstance(node.value, ast.Dict) and len(node.targets) == 1:
target = node.targets[0]
if isinstance(target, ast.Name):
var_name = target.id
dict_values = []
# Extract string values from the dictionary
for val in node.value.values:
if isinstance(val, ast.Constant) and isinstance(val.value, str):
dict_values.append(val.value)
# If non-string, skip or just ignore
if dict_values:
dict_definitions[var_name] = dict_values
self.generic_visit(node)
class ReturnVisitor(ast.NodeVisitor):
def visit_Return(self, node):
# Check if the return value is a constant (Python 3.8+)
if isinstance(node.value, ast.Constant):
return_values.append(node.value.value)
# Direct string return
if isinstance(node.value, ast.Constant) and isinstance(
node.value.value, str
):
return_values.add(node.value.value)
# Dictionary-based return, like return paths[result]
elif isinstance(node.value, ast.Subscript):
# Check if we're subscripting a known dictionary variable
if isinstance(node.value.value, ast.Name):
var_name = node.value.value.id
if var_name in dict_definitions:
# Add all possible dictionary values
for v in dict_definitions[var_name]:
return_values.add(v)
self.generic_visit(node)
# First pass: identify dictionary assignments
DictionaryAssignmentVisitor().visit(code_ast)
# Second pass: identify returns
ReturnVisitor().visit(code_ast)
return return_values
return list(return_values) if return_values else None
def calculate_node_levels(flow):
levels = {}
queue = []
visited = set()
pending_and_listeners = {}
def calculate_node_levels(flow: Any) -> Dict[str, int]:
"""
Calculate the hierarchical level of each node in the flow.
Performs a breadth-first traversal of the flow graph to assign levels
to nodes, starting with start methods at level 0.
Parameters
----------
flow : Any
The flow instance containing methods, listeners, and router configurations.
Returns
-------
Dict[str, int]
Dictionary mapping method names to their hierarchical levels.
Notes
-----
- Start methods are assigned level 0
- Each subsequent connected node is assigned level = parent_level + 1
- Handles both OR and AND conditions for listeners
- Processes router paths separately
"""
levels: Dict[str, int] = {}
queue: List[str] = []
visited: Set[str] = set()
pending_and_listeners: Dict[str, Set[str]] = {}
# Make all start methods at level 0
for method_name, method in flow._methods.items():
@@ -61,10 +134,7 @@ def calculate_node_levels(flow):
current_level = levels[current]
visited.add(current)
for listener_name, (
condition_type,
trigger_methods,
) in flow._listeners.items():
for listener_name, (condition_type, trigger_methods) in flow._listeners.items():
if condition_type == "OR":
if current in trigger_methods:
if (
@@ -89,7 +159,7 @@ def calculate_node_levels(flow):
queue.append(listener_name)
# Handle router connections
if current in flow._routers.values():
if current in flow._routers:
router_method_name = current
paths = flow._router_paths.get(router_method_name, [])
for path in paths:
@@ -105,10 +175,24 @@ def calculate_node_levels(flow):
levels[listener_name] = current_level + 1
if listener_name not in visited:
queue.append(listener_name)
return levels
def count_outgoing_edges(flow):
def count_outgoing_edges(flow: Any) -> Dict[str, int]:
"""
Count the number of outgoing edges for each method in the flow.
Parameters
----------
flow : Any
The flow instance to analyze.
Returns
-------
Dict[str, int]
Dictionary mapping method names to their outgoing edge count.
"""
counts = {}
for method_name in flow._methods:
counts[method_name] = 0
@@ -120,16 +204,53 @@ def count_outgoing_edges(flow):
return counts
def build_ancestor_dict(flow):
ancestors = {node: set() for node in flow._methods}
visited = set()
def build_ancestor_dict(flow: Any) -> Dict[str, Set[str]]:
"""
Build a dictionary mapping each node to its ancestor nodes.
Parameters
----------
flow : Any
The flow instance to analyze.
Returns
-------
Dict[str, Set[str]]
Dictionary mapping each node to a set of its ancestor nodes.
"""
ancestors: Dict[str, Set[str]] = {node: set() for node in flow._methods}
visited: Set[str] = set()
for node in flow._methods:
if node not in visited:
dfs_ancestors(node, ancestors, visited, flow)
return ancestors
def dfs_ancestors(node, ancestors, visited, flow):
def dfs_ancestors(
node: str,
ancestors: Dict[str, Set[str]],
visited: Set[str],
flow: Any
) -> None:
"""
Perform depth-first search to build ancestor relationships.
Parameters
----------
node : str
Current node being processed.
ancestors : Dict[str, Set[str]]
Dictionary tracking ancestor relationships.
visited : Set[str]
Set of already visited nodes.
flow : Any
The flow instance being analyzed.
Notes
-----
This function modifies the ancestors dictionary in-place to build
the complete ancestor graph.
"""
if node in visited:
return
visited.add(node)
@@ -142,7 +263,7 @@ def dfs_ancestors(node, ancestors, visited, flow):
dfs_ancestors(listener_name, ancestors, visited, flow)
# Handle router methods separately
if node in flow._routers.values():
if node in flow._routers:
router_method_name = node
paths = flow._router_paths.get(router_method_name, [])
for path in paths:
@@ -153,12 +274,48 @@ def dfs_ancestors(node, ancestors, visited, flow):
dfs_ancestors(listener_name, ancestors, visited, flow)
def is_ancestor(node, ancestor_candidate, ancestors):
def is_ancestor(node: str, ancestor_candidate: str, ancestors: Dict[str, Set[str]]) -> bool:
"""
Check if one node is an ancestor of another.
Parameters
----------
node : str
The node to check ancestors for.
ancestor_candidate : str
The potential ancestor node.
ancestors : Dict[str, Set[str]]
Dictionary containing ancestor relationships.
Returns
-------
bool
True if ancestor_candidate is an ancestor of node, False otherwise.
"""
return ancestor_candidate in ancestors.get(node, set())
def build_parent_children_dict(flow):
parent_children = {}
def build_parent_children_dict(flow: Any) -> Dict[str, List[str]]:
"""
Build a dictionary mapping parent nodes to their children.
Parameters
----------
flow : Any
The flow instance to analyze.
Returns
-------
Dict[str, List[str]]
Dictionary mapping parent method names to lists of their child method names.
Notes
-----
- Maps listeners to their trigger methods
- Maps router methods to their paths and listeners
- Children lists are sorted for consistent ordering
"""
parent_children: Dict[str, List[str]] = {}
# Map listeners to their trigger methods
for listener_name, (_, trigger_methods) in flow._listeners.items():
@@ -182,7 +339,24 @@ def build_parent_children_dict(flow):
return parent_children
def get_child_index(parent, child, parent_children):
def get_child_index(parent: str, child: str, parent_children: Dict[str, List[str]]) -> int:
"""
Get the index of a child node in its parent's sorted children list.
Parameters
----------
parent : str
The parent node name.
child : str
The child node name to find the index for.
parent_children : Dict[str, List[str]]
Dictionary mapping parents to their children lists.
Returns
-------
int
Zero-based index of the child in its parent's sorted children list.
"""
children = parent_children.get(parent, [])
children.sort()
return children.index(child)

View File

@@ -1,5 +1,23 @@
"""
Utilities for creating visual representations of flow structures.
This module provides functions for generating network visualizations of flows,
including node placement, edge creation, and visual styling. It handles the
conversion of flow structures into visual network graphs with appropriate
styling and layout.
Example
-------
>>> flow = Flow()
>>> net = Network(directed=True)
>>> node_positions = compute_positions(flow, node_levels)
>>> add_nodes_to_network(net, flow, node_positions, node_styles)
>>> add_edges(net, flow, node_positions, colors)
"""
import ast
import inspect
from typing import Any, Dict, List, Optional, Tuple, Union
from .utils import (
build_ancestor_dict,
@@ -9,8 +27,25 @@ from .utils import (
)
def method_calls_crew(method):
"""Check if the method calls `.crew()`."""
def method_calls_crew(method: Any) -> bool:
"""
Check if the method contains a call to `.crew()`.
Parameters
----------
method : Any
The method to analyze for crew() calls.
Returns
-------
bool
True if the method calls .crew(), False otherwise.
Notes
-----
Uses AST analysis to detect method calls, specifically looking for
attribute access of 'crew'.
"""
try:
source = inspect.getsource(method)
source = inspect.cleandoc(source)
@@ -20,6 +55,7 @@ def method_calls_crew(method):
return False
class CrewCallVisitor(ast.NodeVisitor):
"""AST visitor to detect .crew() method calls."""
def __init__(self):
self.found = False
@@ -34,7 +70,34 @@ def method_calls_crew(method):
return visitor.found
def add_nodes_to_network(net, flow, node_positions, node_styles):
def add_nodes_to_network(
net: Any,
flow: Any,
node_positions: Dict[str, Tuple[float, float]],
node_styles: Dict[str, Dict[str, Any]]
) -> None:
"""
Add nodes to the network visualization with appropriate styling.
Parameters
----------
net : Any
The pyvis Network instance to add nodes to.
flow : Any
The flow instance containing method information.
node_positions : Dict[str, Tuple[float, float]]
Dictionary mapping node names to their (x, y) positions.
node_styles : Dict[str, Dict[str, Any]]
Dictionary containing style configurations for different node types.
Notes
-----
Node types include:
- Start methods
- Router methods
- Crew methods
- Regular methods
"""
def human_friendly_label(method_name):
return method_name.replace("_", " ").title()
@@ -73,9 +136,33 @@ def add_nodes_to_network(net, flow, node_positions, node_styles):
)
def compute_positions(flow, node_levels, y_spacing=150, x_spacing=150):
level_nodes = {}
node_positions = {}
def compute_positions(
flow: Any,
node_levels: Dict[str, int],
y_spacing: float = 150,
x_spacing: float = 150
) -> Dict[str, Tuple[float, float]]:
"""
Compute the (x, y) positions for each node in the flow graph.
Parameters
----------
flow : Any
The flow instance to compute positions for.
node_levels : Dict[str, int]
Dictionary mapping node names to their hierarchical levels.
y_spacing : float, optional
Vertical spacing between levels, by default 150.
x_spacing : float, optional
Horizontal spacing between nodes, by default 150.
Returns
-------
Dict[str, Tuple[float, float]]
Dictionary mapping node names to their (x, y) coordinates.
"""
level_nodes: Dict[int, List[str]] = {}
node_positions: Dict[str, Tuple[float, float]] = {}
for method_name, level in node_levels.items():
level_nodes.setdefault(level, []).append(method_name)
@@ -90,16 +177,44 @@ def compute_positions(flow, node_levels, y_spacing=150, x_spacing=150):
return node_positions
def add_edges(net, flow, node_positions, colors):
def add_edges(
net: Any,
flow: Any,
node_positions: Dict[str, Tuple[float, float]],
colors: Dict[str, str]
) -> None:
edge_smooth: Dict[str, Union[str, float]] = {"type": "continuous"} # Default value
"""
Add edges to the network visualization with appropriate styling.
Parameters
----------
net : Any
The pyvis Network instance to add edges to.
flow : Any
The flow instance containing edge information.
node_positions : Dict[str, Tuple[float, float]]
Dictionary mapping node names to their positions.
colors : Dict[str, str]
Dictionary mapping edge types to their colors.
Notes
-----
- Handles both normal listener edges and router edges
- Applies appropriate styling (color, dashes) based on edge type
- Adds curvature to edges when needed (cycles or multiple children)
"""
ancestors = build_ancestor_dict(flow)
parent_children = build_parent_children_dict(flow)
# Edges for normal listeners
for method_name in flow._listeners:
condition_type, trigger_methods = flow._listeners[method_name]
is_and_condition = condition_type == "AND"
for trigger in trigger_methods:
if trigger in flow._methods or trigger in flow._routers.values():
# Check if nodes exist before adding edges
if trigger in node_positions and method_name in node_positions:
is_router_edge = any(
trigger in paths for paths in flow._router_paths.values()
)
@@ -124,7 +239,7 @@ def add_edges(net, flow, node_positions, colors):
else:
edge_smooth = {"type": "cubicBezier"}
else:
edge_smooth = False
edge_smooth.update({"type": "continuous"})
edge_style = {
"color": edge_color,
@@ -135,7 +250,22 @@ def add_edges(net, flow, node_positions, colors):
}
net.add_edge(trigger, method_name, **edge_style)
else:
# Nodes not found in node_positions. Check if it's a known router outcome and a known method.
is_router_edge = any(
trigger in paths for paths in flow._router_paths.values()
)
# Check if method_name is a known method
method_known = method_name in flow._methods
# If it's a known router edge and the method is known, don't warn.
# This means the path is legitimate, just not reflected as nodes here.
if not (is_router_edge and method_known):
print(
f"Warning: No node found for '{trigger}' or '{method_name}'. Skipping edge."
)
# Edges for router return paths
for router_method_name, paths in flow._router_paths.items():
for path in paths:
for listener_name, (
@@ -143,36 +273,49 @@ def add_edges(net, flow, node_positions, colors):
trigger_methods,
) in flow._listeners.items():
if path in trigger_methods:
is_cycle_edge = is_ancestor(trigger, method_name, ancestors)
parent_has_multiple_children = (
len(parent_children.get(router_method_name, [])) > 1
)
needs_curvature = is_cycle_edge or parent_has_multiple_children
if (
router_method_name in node_positions
and listener_name in node_positions
):
is_cycle_edge = is_ancestor(
router_method_name, listener_name, ancestors
)
parent_has_multiple_children = (
len(parent_children.get(router_method_name, [])) > 1
)
needs_curvature = is_cycle_edge or parent_has_multiple_children
if needs_curvature:
source_pos = node_positions.get(router_method_name)
target_pos = node_positions.get(listener_name)
if needs_curvature:
source_pos = node_positions.get(router_method_name)
target_pos = node_positions.get(listener_name)
if source_pos and target_pos:
dx = target_pos[0] - source_pos[0]
smooth_type = "curvedCCW" if dx <= 0 else "curvedCW"
index = get_child_index(
router_method_name, listener_name, parent_children
)
edge_smooth = {
"type": smooth_type,
"roundness": 0.2 + (0.1 * index),
}
if source_pos and target_pos:
dx = target_pos[0] - source_pos[0]
smooth_type = "curvedCCW" if dx <= 0 else "curvedCW"
index = get_child_index(
router_method_name, listener_name, parent_children
)
edge_smooth = {
"type": smooth_type,
"roundness": 0.2 + (0.1 * index),
}
else:
edge_smooth = {"type": "cubicBezier"}
else:
edge_smooth = {"type": "cubicBezier"}
else:
edge_smooth = False
edge_smooth.update({"type": "continuous"})
edge_style = {
"color": colors["router_edge"],
"width": 2,
"arrows": "to",
"dashes": True,
"smooth": edge_smooth,
}
net.add_edge(router_method_name, listener_name, **edge_style)
edge_style = {
"color": colors["router_edge"],
"width": 2,
"arrows": "to",
"dashes": True,
"smooth": edge_smooth,
}
net.add_edge(router_method_name, listener_name, **edge_style)
else:
# Same check here: known router edge and known method?
method_known = listener_name in flow._methods
if not method_known:
print(
f"Warning: No node found for '{router_method_name}' or '{listener_name}'. Skipping edge."
)

View File

@@ -14,13 +14,13 @@ class Knowledge(BaseModel):
Knowledge is a collection of sources and setup for the vector store to save and query relevant context.
Args:
sources: List[BaseKnowledgeSource] = Field(default_factory=list)
storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
storage: Optional[KnowledgeStorage] = Field(default=None)
embedder_config: Optional[Dict[str, Any]] = None
"""
sources: List[BaseKnowledgeSource] = Field(default_factory=list)
model_config = ConfigDict(arbitrary_types_allowed=True)
storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
storage: Optional[KnowledgeStorage] = Field(default=None)
embedder_config: Optional[Dict[str, Any]] = None
collection_name: Optional[str] = None
@@ -49,8 +49,13 @@ class Knowledge(BaseModel):
"""
Query across all knowledge sources to find the most relevant information.
Returns the top_k most relevant chunks.
Raises:
ValueError: If storage is not initialized.
"""
if self.storage is None:
raise ValueError("Storage is not initialized.")
results = self.storage.search(
query,
limit,

View File

@@ -1,8 +1,8 @@
from abc import ABC, abstractmethod
from pathlib import Path
from typing import Dict, List, Union
from typing import Dict, List, Optional, Union
from pydantic import Field
from pydantic import Field, field_validator
from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
from crewai.knowledge.storage.knowledge_storage import KnowledgeStorage
@@ -14,17 +14,29 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
"""Base class for knowledge sources that load content from files."""
_logger: Logger = Logger(verbose=True)
file_path: Union[Path, List[Path], str, List[str]] = Field(
..., description="The path to the file"
file_path: Optional[Union[Path, List[Path], str, List[str]]] = Field(
default=None,
description="[Deprecated] The path to the file. Use file_paths instead.",
)
file_paths: Optional[Union[Path, List[Path], str, List[str]]] = Field(
default_factory=list, description="The path to the file"
)
content: Dict[Path, str] = Field(init=False, default_factory=dict)
storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
storage: Optional[KnowledgeStorage] = Field(default=None)
safe_file_paths: List[Path] = Field(default_factory=list)
@field_validator("file_path", "file_paths", mode="before")
def validate_file_path(cls, v, info):
"""Validate that at least one of file_path or file_paths is provided."""
# Single check if both are None, O(1) instead of nested conditions
if v is None and info.data.get("file_path" if info.field_name == "file_paths" else "file_paths") is None:
raise ValueError("Either file_path or file_paths must be provided")
return v
def model_post_init(self, _):
"""Post-initialization method to load content."""
self.safe_file_paths = self._process_file_paths()
self.validate_paths()
self.validate_content()
self.content = self.load_content()
@abstractmethod
@@ -32,13 +44,13 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
"""Load and preprocess file content. Should be overridden by subclasses. Assume that the file path is relative to the project root in the knowledge directory."""
pass
def validate_paths(self):
def validate_content(self):
"""Validate the paths."""
for path in self.safe_file_paths:
if not path.exists():
self._logger.log(
"error",
f"File not found: {path}. Try adding sources to the knowledge directory. If its inside the knowledge directory, use the relative path.",
f"File not found: {path}. Try adding sources to the knowledge directory. If it's inside the knowledge directory, use the relative path.",
color="red",
)
raise FileNotFoundError(f"File not found: {path}")
@@ -51,7 +63,10 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
def _save_documents(self):
"""Save the documents to the storage."""
self.storage.save(self.chunks)
if self.storage:
self.storage.save(self.chunks)
else:
raise ValueError("No storage found to save documents.")
def convert_to_path(self, path: Union[Path, str]) -> Path:
"""Convert a path to a Path object."""
@@ -59,13 +74,30 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
def _process_file_paths(self) -> List[Path]:
"""Convert file_path to a list of Path objects."""
paths = (
[self.file_path]
if isinstance(self.file_path, (str, Path))
else self.file_path
if hasattr(self, "file_path") and self.file_path is not None:
self._logger.log(
"warning",
"The 'file_path' attribute is deprecated and will be removed in a future version. Please use 'file_paths' instead.",
color="yellow",
)
self.file_paths = self.file_path
if self.file_paths is None:
raise ValueError("Your source must be provided with a file_paths: []")
# Convert single path to list
path_list: List[Union[Path, str]] = (
[self.file_paths]
if isinstance(self.file_paths, (str, Path))
else list(self.file_paths)
if isinstance(self.file_paths, list)
else []
)
if not isinstance(paths, list):
raise ValueError("file_path must be a Path, str, or a list of these types")
if not path_list:
raise ValueError(
"file_path/file_paths must be a Path, str, or a list of these types"
)
return [self.convert_to_path(path) for path in paths]
return [self.convert_to_path(path) for path in path_list]

View File

@@ -16,12 +16,12 @@ class BaseKnowledgeSource(BaseModel, ABC):
chunk_embeddings: List[np.ndarray] = Field(default_factory=list)
model_config = ConfigDict(arbitrary_types_allowed=True)
storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
storage: Optional[KnowledgeStorage] = Field(default=None)
metadata: Dict[str, Any] = Field(default_factory=dict) # Currently unused
collection_name: Optional[str] = Field(default=None)
@abstractmethod
def load_content(self) -> Dict[Any, str]:
def validate_content(self) -> Any:
"""Load and preprocess content from the source."""
pass
@@ -46,4 +46,7 @@ class BaseKnowledgeSource(BaseModel, ABC):
Save the documents to the storage.
This method should be called after the chunks and embeddings are generated.
"""
self.storage.save(self.chunks)
if self.storage:
self.storage.save(self.chunks)
else:
raise ValueError("No storage found to save documents.")

View File

@@ -0,0 +1,120 @@
from pathlib import Path
from typing import Iterator, List, Optional, Union
from urllib.parse import urlparse
from docling.datamodel.base_models import InputFormat
from docling.document_converter import DocumentConverter
from docling.exceptions import ConversionError
from docling_core.transforms.chunker.hierarchical_chunker import HierarchicalChunker
from docling_core.types.doc.document import DoclingDocument
from pydantic import Field
from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
from crewai.utilities.constants import KNOWLEDGE_DIRECTORY
from crewai.utilities.logger import Logger
class CrewDoclingSource(BaseKnowledgeSource):
"""Default Source class for converting documents to markdown or json
This will auto support PDF, DOCX, and TXT, XLSX, Images, and HTML files without any additional dependencies and follows the docling package as the source of truth.
"""
_logger: Logger = Logger(verbose=True)
file_path: Optional[List[Union[Path, str]]] = Field(default=None)
file_paths: List[Union[Path, str]] = Field(default_factory=list)
chunks: List[str] = Field(default_factory=list)
safe_file_paths: List[Union[Path, str]] = Field(default_factory=list)
content: List[DoclingDocument] = Field(default_factory=list)
document_converter: DocumentConverter = Field(
default_factory=lambda: DocumentConverter(
allowed_formats=[
InputFormat.MD,
InputFormat.ASCIIDOC,
InputFormat.PDF,
InputFormat.DOCX,
InputFormat.HTML,
InputFormat.IMAGE,
InputFormat.XLSX,
InputFormat.PPTX,
]
)
)
def model_post_init(self, _) -> None:
if self.file_path:
self._logger.log(
"warning",
"The 'file_path' attribute is deprecated and will be removed in a future version. Please use 'file_paths' instead.",
color="yellow",
)
self.file_paths = self.file_path
self.safe_file_paths = self.validate_content()
self.content = self._load_content()
def _load_content(self) -> List[DoclingDocument]:
try:
return self._convert_source_to_docling_documents()
except ConversionError as e:
self._logger.log(
"error",
f"Error loading content: {e}. Supported formats: {self.document_converter.allowed_formats}",
"red",
)
raise e
except Exception as e:
self._logger.log("error", f"Error loading content: {e}")
raise e
def add(self) -> None:
if self.content is None:
return
for doc in self.content:
new_chunks_iterable = self._chunk_doc(doc)
self.chunks.extend(list(new_chunks_iterable))
self._save_documents()
def _convert_source_to_docling_documents(self) -> List[DoclingDocument]:
conv_results_iter = self.document_converter.convert_all(self.safe_file_paths)
return [result.document for result in conv_results_iter]
def _chunk_doc(self, doc: DoclingDocument) -> Iterator[str]:
chunker = HierarchicalChunker()
for chunk in chunker.chunk(doc):
yield chunk.text
def validate_content(self) -> List[Union[Path, str]]:
processed_paths: List[Union[Path, str]] = []
for path in self.file_paths:
if isinstance(path, str):
if path.startswith(("http://", "https://")):
try:
if self._validate_url(path):
processed_paths.append(path)
else:
raise ValueError(f"Invalid URL format: {path}")
except Exception as e:
raise ValueError(f"Invalid URL: {path}. Error: {str(e)}")
else:
local_path = Path(KNOWLEDGE_DIRECTORY + "/" + path)
if local_path.exists():
processed_paths.append(local_path)
else:
raise FileNotFoundError(f"File not found: {local_path}")
else:
# this is an instance of Path
processed_paths.append(path)
return processed_paths
def _validate_url(self, url: str) -> bool:
try:
result = urlparse(url)
return all(
[
result.scheme in ("http", "https"),
result.netloc,
len(result.netloc.split(".")) >= 2, # Ensure domain has TLD
]
)
except Exception:
return False

View File

@@ -13,9 +13,9 @@ class StringKnowledgeSource(BaseKnowledgeSource):
def model_post_init(self, _):
"""Post-initialization method to validate content."""
self.load_content()
self.validate_content()
def load_content(self):
def validate_content(self):
"""Validate string content."""
if not isinstance(self.content, str):
raise ValueError("StringKnowledgeSource only accepts string content")

View File

@@ -1,5 +1,5 @@
from abc import ABC, abstractmethod
from typing import Dict, Any, List, Optional
from typing import Any, Dict, List, Optional
class BaseKnowledgeStorage(ABC):

View File

@@ -14,9 +14,9 @@ from chromadb.config import Settings
from crewai.knowledge.storage.base_knowledge_storage import BaseKnowledgeStorage
from crewai.utilities import EmbeddingConfigurator
from crewai.utilities.constants import KNOWLEDGE_DIRECTORY
from crewai.utilities.logger import Logger
from crewai.utilities.paths import db_storage_path
from crewai.utilities.constants import KNOWLEDGE_DIRECTORY
@contextlib.contextmanager
@@ -124,43 +124,60 @@ class KnowledgeStorage(BaseKnowledgeStorage):
documents: List[str],
metadata: Optional[Union[Dict[str, Any], List[Dict[str, Any]]]] = None,
):
if self.collection:
try:
if metadata is None:
metadatas: Optional[OneOrMany[chromadb.Metadata]] = None
elif isinstance(metadata, list):
metadatas = [cast(chromadb.Metadata, m) for m in metadata]
else:
metadatas = cast(chromadb.Metadata, metadata)
ids = [
hashlib.sha256(doc.encode("utf-8")).hexdigest() for doc in documents
]
self.collection.upsert(
documents=documents,
metadatas=metadatas,
ids=ids,
)
except chromadb.errors.InvalidDimensionException as e:
Logger(verbose=True).log(
"error",
"Embedding dimension mismatch. This usually happens when mixing different embedding models. Try resetting the collection using `crewai reset-memories -a`",
"red",
)
raise ValueError(
"Embedding dimension mismatch. Make sure you're using the same embedding model "
"across all operations with this collection."
"Try resetting the collection using `crewai reset-memories -a`"
) from e
except Exception as e:
Logger(verbose=True).log(
"error", f"Failed to upsert documents: {e}", "red"
)
raise
else:
if not self.collection:
raise Exception("Collection not initialized")
try:
# Create a dictionary to store unique documents
unique_docs = {}
# Generate IDs and create a mapping of id -> (document, metadata)
for idx, doc in enumerate(documents):
doc_id = hashlib.sha256(doc.encode("utf-8")).hexdigest()
doc_metadata = None
if metadata is not None:
if isinstance(metadata, list):
doc_metadata = metadata[idx]
else:
doc_metadata = metadata
unique_docs[doc_id] = (doc, doc_metadata)
# Prepare filtered lists for ChromaDB
filtered_docs = []
filtered_metadata = []
filtered_ids = []
# Build the filtered lists
for doc_id, (doc, meta) in unique_docs.items():
filtered_docs.append(doc)
filtered_metadata.append(meta)
filtered_ids.append(doc_id)
# If we have no metadata at all, set it to None
final_metadata: Optional[OneOrMany[chromadb.Metadata]] = (
None if all(m is None for m in filtered_metadata) else filtered_metadata
)
self.collection.upsert(
documents=filtered_docs,
metadatas=final_metadata,
ids=filtered_ids,
)
except chromadb.errors.InvalidDimensionException as e:
Logger(verbose=True).log(
"error",
"Embedding dimension mismatch. This usually happens when mixing different embedding models. Try resetting the collection using `crewai reset-memories -a`",
"red",
)
raise ValueError(
"Embedding dimension mismatch. Make sure you're using the same embedding model "
"across all operations with this collection."
"Try resetting the collection using `crewai reset-memories -a`"
) from e
except Exception as e:
Logger(verbose=True).log("error", f"Failed to upsert documents: {e}", "red")
raise
def _create_default_embedding_function(self):
from chromadb.utils.embedding_functions.openai_embedding_function import (
OpenAIEmbeddingFunction,

View File

@@ -43,6 +43,11 @@ LLM_CONTEXT_WINDOW_SIZES = {
"gpt-4-turbo": 128000,
"o1-preview": 128000,
"o1-mini": 128000,
# gemini
"gemini-2.0-flash": 1048576,
"gemini-1.5-pro": 2097152,
"gemini-1.5-flash": 1048576,
"gemini-1.5-flash-8b": 1048576,
# deepseek
"deepseek-chat": 128000,
# groq
@@ -59,8 +64,13 @@ LLM_CONTEXT_WINDOW_SIZES = {
"llama3-70b-8192": 8192,
"llama3-8b-8192": 8192,
"mixtral-8x7b-32768": 32768,
"llama-3.3-70b-versatile": 128000,
"llama-3.3-70b-instruct": 128000,
}
DEFAULT_CONTEXT_WINDOW_SIZE = 8192
CONTEXT_WINDOW_USAGE_RATIO = 0.75
@contextmanager
def suppress_warnings():
@@ -124,6 +134,7 @@ class LLM:
self.api_version = api_version
self.api_key = api_key
self.callbacks = callbacks
self.context_window_size = 0
self.kwargs = kwargs
litellm.drop_params = True
@@ -191,7 +202,16 @@ class LLM:
def get_context_window_size(self) -> int:
# Only using 75% of the context window size to avoid cutting the message in the middle
return int(LLM_CONTEXT_WINDOW_SIZES.get(self.model, 8192) * 0.75)
if self.context_window_size != 0:
return self.context_window_size
self.context_window_size = int(
DEFAULT_CONTEXT_WINDOW_SIZE * CONTEXT_WINDOW_USAGE_RATIO
)
for key, value in LLM_CONTEXT_WINDOW_SIZES.items():
if self.model.startswith(key):
self.context_window_size = int(value * CONTEXT_WINDOW_USAGE_RATIO)
return self.context_window_size
def set_callbacks(self, callbacks: List[Any]):
callback_types = [type(callback) for callback in callbacks]

View File

@@ -1,4 +1,4 @@
from typing import Optional, Dict, Any
from typing import Any, Dict, Optional
from crewai.memory import EntityMemory, LongTermMemory, ShortTermMemory, UserMemory

View File

@@ -1,4 +1,4 @@
from typing import Any, Dict, Optional, List
from typing import Any, Dict, List, Optional
from crewai.memory.storage.rag_storage import RAGStorage

View File

@@ -1,4 +1,5 @@
from typing import Any, Dict, Optional
from crewai.memory.memory import Memory
from crewai.memory.short_term.short_term_memory_item import ShortTermMemoryItem
from crewai.memory.storage.rag_storage import RAGStorage
@@ -32,7 +33,10 @@ class ShortTermMemory(Memory):
storage
if storage
else RAGStorage(
type="short_term", embedder_config=embedder_config, crew=crew, path=path
type="short_term",
embedder_config=embedder_config,
crew=crew,
path=path,
)
)
super().__init__(storage)

View File

@@ -2,6 +2,7 @@ import os
from typing import Any, Dict, List
from mem0 import MemoryClient
from crewai.memory.storage.interface import Storage

View File

@@ -1,5 +1,6 @@
from functools import wraps
def memoize(func):
cache = {}

View File

@@ -1,12 +1,25 @@
import datetime
import inspect
import json
from pathlib import Path
import logging
import threading
import uuid
from concurrent.futures import Future
from copy import copy
from hashlib import md5
from typing import Any, Dict, List, Optional, Set, Tuple, Type, Union
from pathlib import Path
from typing import (
Any,
Callable,
ClassVar,
Dict,
List,
Optional,
Set,
Tuple,
Type,
Union,
)
from opentelemetry.trace import Span
from pydantic import (
@@ -20,6 +33,7 @@ from pydantic import (
from pydantic_core import PydanticCustomError
from crewai.agents.agent_builder.base_agent import BaseAgent
from crewai.tasks.guardrail_result import GuardrailResult
from crewai.tasks.output_format import OutputFormat
from crewai.tasks.task_output import TaskOutput
from crewai.telemetry.telemetry import Telemetry
@@ -49,6 +63,7 @@ class Task(BaseModel):
"""
__hash__ = object.__hash__ # type: ignore
logger: ClassVar[logging.Logger] = logging.getLogger(__name__)
used_tools: int = 0
tools_errors: int = 0
delegations: int = 0
@@ -110,11 +125,61 @@ class Task(BaseModel):
default=None,
)
processed_by_agents: Set[str] = Field(default_factory=set)
guardrail: Optional[Callable[[TaskOutput], Tuple[bool, Any]]] = Field(
default=None,
description="Function to validate task output before proceeding to next task"
)
max_retries: int = Field(
default=3,
description="Maximum number of retries when guardrail fails"
)
retry_count: int = Field(
default=0,
description="Current number of retries"
)
@field_validator("guardrail")
@classmethod
def validate_guardrail_function(cls, v: Optional[Callable]) -> Optional[Callable]:
"""Validate that the guardrail function has the correct signature and behavior.
While type hints provide static checking, this validator ensures runtime safety by:
1. Verifying the function accepts exactly one parameter (the TaskOutput)
2. Checking return type annotations match Tuple[bool, Any] if present
3. Providing clear, immediate error messages for debugging
This runtime validation is crucial because:
- Type hints are optional and can be ignored at runtime
- Function signatures need immediate validation before task execution
- Clear error messages help users debug guardrail implementation issues
Args:
v: The guardrail function to validate
Returns:
The validated guardrail function
Raises:
ValueError: If the function signature is invalid or return annotation
doesn't match Tuple[bool, Any]
"""
if v is not None:
sig = inspect.signature(v)
if len(sig.parameters) != 1:
raise ValueError("Guardrail function must accept exactly one parameter")
# Check return annotation if present, but don't require it
return_annotation = sig.return_annotation
if return_annotation != inspect.Signature.empty:
if not (return_annotation == Tuple[bool, Any] or str(return_annotation) == 'Tuple[bool, Any]'):
raise ValueError("If return type is annotated, it must be Tuple[bool, Any]")
return v
_telemetry: Telemetry = PrivateAttr(default_factory=Telemetry)
_execution_span: Optional[Span] = PrivateAttr(default=None)
_original_description: Optional[str] = PrivateAttr(default=None)
_original_expected_output: Optional[str] = PrivateAttr(default=None)
_original_output_file: Optional[str] = PrivateAttr(default=None)
_thread: Optional[threading.Thread] = PrivateAttr(default=None)
_execution_time: Optional[float] = PrivateAttr(default=None)
@@ -149,8 +214,46 @@ class Task(BaseModel):
@field_validator("output_file")
@classmethod
def output_file_validation(cls, value: str) -> str:
"""Validate the output file path by removing the / from the beginning of the path."""
def output_file_validation(cls, value: Optional[str]) -> Optional[str]:
"""Validate the output file path.
Args:
value: The output file path to validate. Can be None or a string.
If the path contains template variables (e.g. {var}), leading slashes are preserved.
For regular paths, leading slashes are stripped.
Returns:
The validated and potentially modified path, or None if no path was provided.
Raises:
ValueError: If the path contains invalid characters, path traversal attempts,
or other security concerns.
"""
if value is None:
return None
# Basic security checks
if ".." in value:
raise ValueError("Path traversal attempts are not allowed in output_file paths")
# Check for shell expansion first
if value.startswith('~') or value.startswith('$'):
raise ValueError("Shell expansion characters are not allowed in output_file paths")
# Then check other shell special characters
if any(char in value for char in ['|', '>', '<', '&', ';']):
raise ValueError("Shell special characters are not allowed in output_file paths")
# Don't strip leading slash if it's a template path with variables
if "{" in value or "}" in value:
# Validate template variable format
template_vars = [part.split("}")[0] for part in value.split("{")[1:]]
for var in template_vars:
if not var.isidentifier():
raise ValueError(f"Invalid template variable name: {var}")
return value
# Strip leading slash for regular paths
if value.startswith("/"):
return value[1:]
return value
@@ -254,7 +357,6 @@ class Task(BaseModel):
)
pydantic_output, json_output = self._export_output(result)
task_output = TaskOutput(
name=self.name,
description=self.description,
@@ -265,6 +367,37 @@ class Task(BaseModel):
agent=agent.role,
output_format=self._get_output_format(),
)
if self.guardrail:
guardrail_result = GuardrailResult.from_tuple(self.guardrail(task_output))
if not guardrail_result.success:
if self.retry_count >= self.max_retries:
raise Exception(
f"Task failed guardrail validation after {self.max_retries} retries. "
f"Last error: {guardrail_result.error}"
)
self.retry_count += 1
context = (
f"### Previous attempt failed validation: {guardrail_result.error}\n\n\n"
f"### Previous result:\n{task_output.raw}\n\n\n"
"Try again, making sure to address the validation error."
)
return self._execute_core(agent, context, tools)
if guardrail_result.result is None:
raise Exception(
"Task guardrail returned None as result. This is not allowed."
)
if isinstance(guardrail_result.result, str):
task_output.raw = guardrail_result.result
pydantic_output, json_output = self._export_output(guardrail_result.result)
task_output.pydantic = pydantic_output
task_output.json_dict = json_output
elif isinstance(guardrail_result.result, TaskOutput):
task_output = guardrail_result.result
self.output = task_output
self._set_end_execution_time(start_time)
@@ -299,16 +432,89 @@ class Task(BaseModel):
tasks_slices = [self.description, output]
return "\n".join(tasks_slices)
def interpolate_inputs(self, inputs: Dict[str, Any]) -> None:
"""Interpolate inputs into the task description and expected output."""
def interpolate_inputs(self, inputs: Dict[str, Union[str, int, float]]) -> None:
"""Interpolate inputs into the task description, expected output, and output file path.
Args:
inputs: Dictionary mapping template variables to their values.
Supported value types are strings, integers, and floats.
Raises:
ValueError: If a required template variable is missing from inputs.
"""
if self._original_description is None:
self._original_description = self.description
if self._original_expected_output is None:
self._original_expected_output = self.expected_output
if self.output_file is not None and self._original_output_file is None:
self._original_output_file = self.output_file
if inputs:
if not inputs:
return
try:
self.description = self._original_description.format(**inputs)
self.expected_output = self._original_expected_output.format(**inputs)
except KeyError as e:
raise ValueError(f"Missing required template variable '{e.args[0]}' in description") from e
except ValueError as e:
raise ValueError(f"Error interpolating description: {str(e)}") from e
try:
self.expected_output = self.interpolate_only(
input_string=self._original_expected_output, inputs=inputs
)
except (KeyError, ValueError) as e:
raise ValueError(f"Error interpolating expected_output: {str(e)}") from e
if self.output_file is not None:
try:
self.output_file = self.interpolate_only(
input_string=self._original_output_file, inputs=inputs
)
except (KeyError, ValueError) as e:
raise ValueError(f"Error interpolating output_file path: {str(e)}") from e
def interpolate_only(self, input_string: Optional[str], inputs: Dict[str, Union[str, int, float]]) -> str:
"""Interpolate placeholders (e.g., {key}) in a string while leaving JSON untouched.
Args:
input_string: The string containing template variables to interpolate.
Can be None or empty, in which case an empty string is returned.
inputs: Dictionary mapping template variables to their values.
Supported value types are strings, integers, and floats.
If input_string is empty or has no placeholders, inputs can be empty.
Returns:
The interpolated string with all template variables replaced with their values.
Empty string if input_string is None or empty.
Raises:
ValueError: If a required template variable is missing from inputs.
KeyError: If a template variable is not found in the inputs dictionary.
"""
if input_string is None or not input_string:
return ""
if "{" not in input_string and "}" not in input_string:
return input_string
if not inputs:
raise ValueError("Inputs dictionary cannot be empty when interpolating variables")
try:
# Validate input types
for key, value in inputs.items():
if not isinstance(value, (str, int, float)):
raise ValueError(f"Value for key '{key}' must be a string, integer, or float, got {type(value).__name__}")
escaped_string = input_string.replace("{", "{{").replace("}", "}}")
for key in inputs.keys():
escaped_string = escaped_string.replace(f"{{{{{key}}}}}", f"{{{key}}}")
return escaped_string.format(**inputs)
except KeyError as e:
raise KeyError(f"Template variable '{e.args[0]}' not found in inputs dictionary") from e
except ValueError as e:
raise ValueError(f"Error during string interpolation: {str(e)}") from e
def increment_tools_errors(self) -> None:
"""Increment the tools errors counter."""
@@ -390,22 +596,33 @@ class Task(BaseModel):
return OutputFormat.RAW
def _save_file(self, result: Any) -> None:
"""Save task output to a file.
Args:
result: The result to save to the file. Can be a dict or any stringifiable object.
Raises:
ValueError: If output_file is not set
RuntimeError: If there is an error writing to the file
"""
if self.output_file is None:
raise ValueError("output_file is not set.")
resolved_path = Path(self.output_file).expanduser().resolve()
directory = resolved_path.parent
try:
resolved_path = Path(self.output_file).expanduser().resolve()
directory = resolved_path.parent
if not directory.exists():
directory.mkdir(parents=True, exist_ok=True)
if not directory.exists():
directory.mkdir(parents=True, exist_ok=True)
with resolved_path.open("w", encoding="utf-8") as file:
if isinstance(result, dict):
import json
json.dump(result, file, ensure_ascii=False, indent=2)
else:
file.write(str(result))
with resolved_path.open("w", encoding="utf-8") as file:
if isinstance(result, dict):
import json
json.dump(result, file, ensure_ascii=False, indent=2)
else:
file.write(str(result))
except (OSError, IOError) as e:
raise RuntimeError(f"Failed to save output file: {e}")
return None
def __repr__(self):

View File

@@ -0,0 +1,56 @@
"""
Module for handling task guardrail validation results.
This module provides the GuardrailResult class which standardizes
the way task guardrails return their validation results.
"""
from typing import Any, Optional, Tuple, Union
from pydantic import BaseModel, field_validator
class GuardrailResult(BaseModel):
"""Result from a task guardrail execution.
This class standardizes the return format of task guardrails,
converting tuple responses into a structured format that can
be easily handled by the task execution system.
Attributes:
success (bool): Whether the guardrail validation passed
result (Any, optional): The validated/transformed result if successful
error (str, optional): Error message if validation failed
"""
success: bool
result: Optional[Any] = None
error: Optional[str] = None
@field_validator("result", "error")
@classmethod
def validate_result_error_exclusivity(cls, v: Any, info) -> Any:
values = info.data
if "success" in values:
if values["success"] and v and "error" in values and values["error"]:
raise ValueError("Cannot have both result and error when success is True")
if not values["success"] and v and "result" in values and values["result"]:
raise ValueError("Cannot have both result and error when success is False")
return v
@classmethod
def from_tuple(cls, result: Tuple[bool, Union[Any, str]]) -> "GuardrailResult":
"""Create a GuardrailResult from a validation tuple.
Args:
result: A tuple of (success, data) where data is either
the validated result or error message.
Returns:
GuardrailResult: A new instance with the tuple data.
"""
success, data = result
return cls(
success=success,
result=data if success else None,
error=data if not success else None
)

View File

@@ -6,6 +6,7 @@ import os
import platform
import warnings
from contextlib import contextmanager
from importlib.metadata import version
from typing import TYPE_CHECKING, Any, Optional
@@ -16,12 +17,10 @@ def suppress_warnings():
yield
with suppress_warnings():
import pkg_resources
from opentelemetry import trace # noqa: E402
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter # noqa: E402
from opentelemetry.exporter.otlp.proto.http.trace_exporter import (
OTLPSpanExporter, # noqa: E402
)
from opentelemetry.sdk.resources import SERVICE_NAME, Resource # noqa: E402
from opentelemetry.sdk.trace import TracerProvider # noqa: E402
from opentelemetry.sdk.trace.export import BatchSpanProcessor # noqa: E402
@@ -104,7 +103,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(span, "python_version", platform.python_version())
self._add_attribute(span, "crew_key", crew.key)
@@ -306,7 +305,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(span, "tool_name", tool_name)
self._add_attribute(span, "attempts", attempts)
@@ -326,7 +325,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(span, "tool_name", tool_name)
self._add_attribute(span, "attempts", attempts)
@@ -346,7 +345,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
if llm:
self._add_attribute(span, "llm", llm.model)
@@ -365,7 +364,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(span, "crew_key", crew.key)
self._add_attribute(span, "crew_id", str(crew.id))
@@ -391,7 +390,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(span, "crew_key", crew.key)
self._add_attribute(span, "crew_id", str(crew.id))
@@ -472,7 +471,7 @@ class Telemetry:
self._add_attribute(
span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(span, "crew_key", crew.key)
self._add_attribute(span, "crew_id", str(crew.id))
@@ -541,7 +540,7 @@ class Telemetry:
self._add_attribute(
crew._execution_span,
"crewai_version",
pkg_resources.get_distribution("crewai").version,
version("crewai"),
)
self._add_attribute(
crew._execution_span, "crew_output", final_string_output

View File

@@ -0,0 +1,45 @@
from typing import Dict, Optional, Union
from pydantic import BaseModel, Field
from crewai.tools.base_tool import BaseTool
from crewai.utilities import I18N
i18n = I18N()
class AddImageToolSchema(BaseModel):
image_url: str = Field(..., description="The URL or path of the image to add")
action: Optional[str] = Field(
default=None,
description="Optional context or question about the image"
)
class AddImageTool(BaseTool):
"""Tool for adding images to the content"""
name: str = Field(default_factory=lambda: i18n.tools("add_image")["name"]) # type: ignore
description: str = Field(default_factory=lambda: i18n.tools("add_image")["description"]) # type: ignore
args_schema: type[BaseModel] = AddImageToolSchema
def _run(
self,
image_url: str,
action: Optional[str] = None,
**kwargs,
) -> dict:
action = action or i18n.tools("add_image")["default_action"] # type: ignore
content = [
{"type": "text", "text": action},
{
"type": "image_url",
"image_url": {
"url": image_url,
},
}
]
return {
"role": "user",
"content": content
}

View File

@@ -1,9 +1,9 @@
from crewai.tools.base_tool import BaseTool
from crewai.agents.agent_builder.base_agent import BaseAgent
from crewai.tools.base_tool import BaseTool
from crewai.utilities import I18N
from .delegate_work_tool import DelegateWorkTool
from .ask_question_tool import AskQuestionTool
from .delegate_work_tool import DelegateWorkTool
class AgentTools:
@@ -20,13 +20,13 @@ class AgentTools:
delegate_tool = DelegateWorkTool(
agents=self.agents,
i18n=self.i18n,
description=self.i18n.tools("delegate_work").format(coworkers=coworkers),
description=self.i18n.tools("delegate_work").format(coworkers=coworkers), # type: ignore
)
ask_tool = AskQuestionTool(
agents=self.agents,
i18n=self.i18n,
description=self.i18n.tools("ask_question").format(coworkers=coworkers),
description=self.i18n.tools("ask_question").format(coworkers=coworkers), # type: ignore
)
return [delegate_tool, ask_tool]

View File

@@ -1,7 +1,9 @@
from crewai.tools.agent_tools.base_agent_tools import BaseAgentTool
from typing import Optional
from pydantic import BaseModel, Field
from crewai.tools.agent_tools.base_agent_tools import BaseAgentTool
class AskQuestionToolSchema(BaseModel):
question: str = Field(..., description="The question to ask")

View File

@@ -1,11 +1,15 @@
import logging
from typing import Optional, Union
from pydantic import Field
from crewai.tools.base_tool import BaseTool
from crewai.agents.agent_builder.base_agent import BaseAgent
from crewai.task import Task
from crewai.tools.base_tool import BaseTool
from crewai.utilities import I18N
logger = logging.getLogger(__name__)
class BaseAgentTool(BaseTool):
"""Base class for agent-related tools"""
@@ -15,6 +19,25 @@ class BaseAgentTool(BaseTool):
default_factory=I18N, description="Internationalization settings"
)
def sanitize_agent_name(self, name: str) -> str:
"""
Sanitize agent role name by normalizing whitespace and setting to lowercase.
Converts all whitespace (including newlines) to single spaces and removes quotes.
Args:
name (str): The agent role name to sanitize
Returns:
str: The sanitized agent role name, with whitespace normalized,
converted to lowercase, and quotes removed
"""
if not name:
return ""
# Normalize all whitespace (including newlines) to single spaces
normalized = " ".join(name.split())
# Remove quotes and convert to lowercase
return normalized.replace('"', "").casefold()
def _get_coworker(self, coworker: Optional[str], **kwargs) -> Optional[str]:
coworker = coworker or kwargs.get("co_worker") or kwargs.get("coworker")
if coworker:
@@ -24,11 +47,27 @@ class BaseAgentTool(BaseTool):
return coworker
def _execute(
self, agent_name: Union[str, None], task: str, context: Union[str, None]
self,
agent_name: Optional[str],
task: str,
context: Optional[str] = None
) -> str:
"""
Execute delegation to an agent with case-insensitive and whitespace-tolerant matching.
Args:
agent_name: Name/role of the agent to delegate to (case-insensitive)
task: The specific question or task to delegate
context: Optional additional context for the task execution
Returns:
str: The execution result from the delegated agent or an error message
if the agent cannot be found
"""
try:
if agent_name is None:
agent_name = ""
logger.debug("No agent name provided, using empty string")
# It is important to remove the quotes from the agent name.
# The reason we have to do this is because less-powerful LLM's
@@ -37,31 +76,49 @@ class BaseAgentTool(BaseTool):
# {"task": "....", "coworker": "....
# when it should look like this:
# {"task": "....", "coworker": "...."}
agent_name = agent_name.casefold().replace('"', "").replace("\n", "")
sanitized_name = self.sanitize_agent_name(agent_name)
logger.debug(f"Sanitized agent name from '{agent_name}' to '{sanitized_name}'")
available_agents = [agent.role for agent in self.agents]
logger.debug(f"Available agents: {available_agents}")
agent = [ # type: ignore # Incompatible types in assignment (expression has type "list[BaseAgent]", variable has type "str | None")
available_agent
for available_agent in self.agents
if available_agent.role.casefold().replace("\n", "") == agent_name
if self.sanitize_agent_name(available_agent.role) == sanitized_name
]
except Exception as _:
return self.i18n.errors("agent_tool_unexsiting_coworker").format(
logger.debug(f"Found {len(agent)} matching agents for role '{sanitized_name}'")
except (AttributeError, ValueError) as e:
# Handle specific exceptions that might occur during role name processing
return self.i18n.errors("agent_tool_unexisting_coworker").format(
coworkers="\n".join(
[f"- {agent.role.casefold()}" for agent in self.agents]
)
[f"- {self.sanitize_agent_name(agent.role)}" for agent in self.agents]
),
error=str(e)
)
if not agent:
return self.i18n.errors("agent_tool_unexsiting_coworker").format(
# No matching agent found after sanitization
return self.i18n.errors("agent_tool_unexisting_coworker").format(
coworkers="\n".join(
[f"- {agent.role.casefold()}" for agent in self.agents]
)
[f"- {self.sanitize_agent_name(agent.role)}" for agent in self.agents]
),
error=f"No agent found with role '{sanitized_name}'"
)
agent = agent[0]
task_with_assigned_agent = Task( # type: ignore # Incompatible types in assignment (expression has type "Task", variable has type "str")
description=task,
agent=agent,
expected_output=agent.i18n.slice("manager_request"),
i18n=agent.i18n,
)
return agent.execute_task(task_with_assigned_agent, context)
try:
task_with_assigned_agent = Task(
description=task,
agent=agent,
expected_output=agent.i18n.slice("manager_request"),
i18n=agent.i18n,
)
logger.debug(f"Created task for agent '{self.sanitize_agent_name(agent.role)}': {task}")
return agent.execute_task(task_with_assigned_agent, context)
except Exception as e:
# Handle task creation or execution errors
return self.i18n.errors("agent_tool_execution_error").format(
agent_role=self.sanitize_agent_name(agent.role),
error=str(e)
)

View File

@@ -1,8 +1,9 @@
from crewai.tools.agent_tools.base_agent_tools import BaseAgentTool
from typing import Optional
from pydantic import BaseModel, Field
from crewai.tools.agent_tools.base_agent_tools import BaseAgentTool
class DelegateWorkToolSchema(BaseModel):
task: str = Field(..., description="The task to delegate")

View File

@@ -1,6 +1,5 @@
import ast
import datetime
import os
import time
from difflib import SequenceMatcher
from textwrap import dedent
@@ -11,18 +10,16 @@ from crewai.agents.tools_handler import ToolsHandler
from crewai.task import Task
from crewai.telemetry import Telemetry
from crewai.tools import BaseTool
from crewai.tools.structured_tool import CrewStructuredTool
from crewai.tools.tool_calling import InstructorToolCalling, ToolCalling
from crewai.tools.tool_usage_events import ToolUsageError, ToolUsageFinished
from crewai.utilities import I18N, Converter, ConverterError, Printer
agentops = None
if os.environ.get("AGENTOPS_API_KEY"):
try:
import agentops # type: ignore
except ImportError:
pass
OPENAI_BIGGER_MODELS = ["gpt-4", "gpt-4o", "o1-preview", "o1-mini"]
try:
import agentops # type: ignore
except ImportError:
agentops = None
OPENAI_BIGGER_MODELS = ["gpt-4", "gpt-4o", "o1-preview", "o1-mini", "o1", "o3", "o3-mini"]
class ToolUsageErrorException(Exception):
@@ -106,6 +103,19 @@ class ToolUsage:
if self.agent.verbose:
self._printer.print(content=f"\n\n{error}\n", color="red")
return error
if isinstance(tool, CrewStructuredTool) and tool.name == self._i18n.tools("add_image")["name"]: # type: ignore
try:
result = self._use(tool_string=tool_string, tool=tool, calling=calling)
return result
except Exception as e:
error = getattr(e, "message", str(e))
self.task.increment_tools_errors()
if self.agent.verbose:
self._printer.print(content=f"\n\n{error}\n", color="red")
return error
return f"{self._use(tool_string=tool_string, tool=tool, calling=calling)}" # type: ignore # BUG?: "_use" of "ToolUsage" does not return a value (it only ever returns None)
def _use(
@@ -422,9 +432,10 @@ class ToolUsage:
elif value.lower() in [
"true",
"false",
"null",
]: # Check for boolean and null values
value = value.lower()
value = value.lower().capitalize()
elif value.lower() == "null":
value = "None"
else:
# Assume the value is a string and needs quotes
value = '"' + value.replace('"', '\\"') + '"'

View File

@@ -1,6 +1,7 @@
from typing import Any, Dict
from pydantic import BaseModel
from datetime import datetime
from typing import Any, Dict
from pydantic import BaseModel
class ToolUsageEvent(BaseModel):

View File

@@ -12,7 +12,7 @@
"tools": "\nYou ONLY have access to the following tools, and should NEVER make up tools that are not listed here:\n\n{tools}\n\nUse the following format:\n\nThought: you should always think about what to do\nAction: the action to take, only one name of [{tool_names}], just the name, exactly as it's written.\nAction Input: the input to the action, just a simple python dictionary, enclosed in curly braces, using \" to wrap keys and values.\nObservation: the result of the action\n\nOnce all necessary information is gathered:\n\nThought: I now know the final answer\nFinal Answer: the final answer to the original input question\n",
"no_tools": "\nTo give my best complete final answer to the task use the exact following format:\n\nThought: I now can give a great answer\nFinal Answer: Your final answer must be the great and the most complete as possible, it must be outcome described.\n\nI MUST use these formats, my job depends on it!",
"format": "I MUST either use a tool (use one at time) OR give my best final answer not both at the same time. To Use the following format:\n\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action, dictionary enclosed in curly braces\nObservation: the result of the action\n... (this Thought/Action/Action Input/Result can repeat N times)\nThought: I now can give a great answer\nFinal Answer: Your final answer must be the great and the most complete as possible, it must be outcome described\n\n",
"final_answer_format": "If you don't need to use any more tools, you must give your best complete final answer, make sure it satisfy the expect criteria, use the EXACT format below:\n\nThought: I now can give a great answer\nFinal Answer: my best complete final answer to the task.\n\n",
"final_answer_format": "If you don't need to use any more tools, you must give your best complete final answer, make sure it satisfies the expected criteria, use the EXACT format below:\n\nThought: I now can give a great answer\nFinal Answer: my best complete final answer to the task.\n\n",
"format_without_tools": "\nSorry, I didn't use the right format. I MUST either use a tool (among the available ones), OR give my best final answer.\nI just remembered the expected format I must follow:\n\nQuestion: the input question you must answer\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action\nObservation: the result of the action\n... (this Thought/Action/Action Input/Result can repeat N times)\nThought: I now can give a great answer\nFinal Answer: Your final answer must be the great and the most complete as possible, it must be outcome described\n\n",
"task_with_context": "{task}\n\nThis is the context you're working with:\n{context}",
"expected_output": "\nThis is the expect criteria for your final answer: {expected_output}\nyou MUST return the actual complete content as the final answer, not a summary.",
@@ -28,15 +28,21 @@
"errors": {
"force_final_answer_error": "You can't keep going, this was the best you could do.\n {formatted_answer.text}",
"force_final_answer": "Now it's time you MUST give your absolute best final answer. You'll ignore all previous instructions, stop using any tools, and just return your absolute BEST Final answer.",
"agent_tool_unexsiting_coworker": "\nError executing tool. coworker mentioned not found, it must be one of the following options:\n{coworkers}\n",
"agent_tool_unexisting_coworker": "\nError executing tool. coworker mentioned not found, it must be one of the following options:\n{coworkers}\n",
"task_repeated_usage": "I tried reusing the same input, I must stop using this action input. I'll try something else instead.\n\n",
"tool_usage_error": "I encountered an error: {error}",
"tool_arguments_error": "Error: the Action Input is not a valid key, value dictionary.",
"wrong_tool_name": "You tried to use the tool {tool}, but it doesn't exist. You must use one of the following tools, use one at time: {tools}.",
"tool_usage_exception": "I encountered an error while trying to use the tool. This was the error: {error}.\n Tool {tool} accepts these inputs: {tool_inputs}"
"tool_usage_exception": "I encountered an error while trying to use the tool. This was the error: {error}.\n Tool {tool} accepts these inputs: {tool_inputs}",
"agent_tool_execution_error": "Error executing task with agent '{agent_role}'. Error: {error}"
},
"tools": {
"delegate_work": "Delegate a specific task to one of the following coworkers: {coworkers}\nThe input to this tool should be the coworker, the task you want them to do, and ALL necessary context to execute the task, they know nothing about the task, so share absolute everything you know, don't reference things but instead explain them.",
"ask_question": "Ask a specific question to one of the following coworkers: {coworkers}\nThe input to this tool should be the coworker, the question you have for them, and ALL necessary context to ask the question properly, they know nothing about the question, so share absolute everything you know, don't reference things but instead explain them."
"ask_question": "Ask a specific question to one of the following coworkers: {coworkers}\nThe input to this tool should be the coworker, the question you have for them, and ALL necessary context to ask the question properly, they know nothing about the question, so share absolute everything you know, don't reference things but instead explain them.",
"add_image": {
"name": "Add image to content",
"description": "See image to understand it's content, you can optionally ask a question about the image",
"default_action": "Please provide a detailed description of this image, including all visual elements, context, and any notable details you can observe."
}
}
}

Some files were not shown because too many files have changed in this diff Show More