Commit Graph

151 Commits

Author SHA1 Message Date
João Moura
b0e38c2a2d adding meomization to crewai project annotations 2024-05-03 00:49:37 -03:00
João Moura
68e1444ed8 updating manager llm pydantic error 2024-05-02 23:39:56 -03:00
João Moura
9648254ec7 Better json parsing for smaller models 2024-05-02 21:57:41 -03:00
David Solito
e280e4a62a Update crew.py (#551)
Ad manager_agent description in crew docstring
2024-05-02 19:21:22 -03:00
João Moura
cbe53d9daf small improvements for i18n 2024-05-02 05:00:29 -03:00
João Moura
7ad0357eaa adding initial support for external prompt file 2024-05-02 05:00:29 -03:00
Jason Schrader
b7d4c4843d fix typos in generated readme (#345)
small things I noticed while upgrading our setup!
2024-05-02 03:32:18 -03:00
Ikko Eltociear Ashimine
93dc0874b5 fix typo (#489)
* Update test_crew_function_calling_llm.yaml

ouput -> output

* Update tool_usage.py

ouput -> output
2024-05-02 03:27:40 -03:00
ftoppi
19fe093ce9 Update task.py: try to find json in task output using regex (#491)
* Update task.py: try to find json in task output using regex

Sometimes the model replies with a valid and additional text, let's try to extract and validate it first. It's cheaper than calling LLM for that.

* Update task.py

---------

Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-05-02 03:26:34 -03:00
Samuel Kocúr
75b495c544 fix db_storage_path handling to use env variable or cwd (#507) 2024-05-02 03:16:54 -03:00
deadlious
e16b1f4a53 Tool name recognition based on string distance (#521)
* adding variations of ask question and delegate work tools

* Revert "adding variations of ask question and delegate work tools"

This reverts commit 38d4589be8.

* adding distance calculation for tool names.

* proper formatting

* remove brackets
2024-05-02 03:15:34 -03:00
Victor Carvalho Tavernari
3f16686627 Add conditional check for output file directory creation (#523)
This commit adds a conditional check to ensure that the output file directory exists before attempting to create it. This ensures that the code does not
fail in cases where the directory does not exist and needs to be created. The condition is added in the `_save_file` method of the `Task` class, ensuring
that the correct behavior is maintained for saving results to a file.
2024-05-02 03:13:51 -03:00
João Moura
2ae6fc4bd8 smal fixes and better guardrail for parsing small models tools usage 2024-05-02 02:21:59 -03:00
João Moura
84775373a6 Adding support for system, prompt and answe templates 2024-05-02 02:21:59 -03:00
João Moura
9d4dc1a081 unifying co-worker string 2024-05-02 02:21:59 -03:00
João Moura
6840fd8ffc remving blank line 2024-05-02 02:21:59 -03:00
João Moura
83596c1a32 Fixing task callback 2024-05-02 02:21:59 -03:00
João Moura
55fcda758a Revert "AgentOps Implementation (#411)"
This reverts commit bf436f885e.
2024-05-02 02:21:59 -03:00
Braelyn Boynton
bf436f885e AgentOps Implementation (#411)
* implements agentops with a langchain handler, agent tracking and tool call recording

* track tool usage

* end session after completion

* track tool usage time

* better tool and llm tracking

* code cleanup

* make agentops optional

* optional dependency usage

* remove telemetry code

* optional agentops

* agentops version bump

* remove org key

* true dependency

* add crew org key to agentops

* cleanup

* Update pyproject.toml

* Revert "true dependency"

This reverts commit e52e8e9568.

* Revert "cleanup"

This reverts commit 7f5635fb9e.

* optional parent key

* agentops 0.1.5

* Revert "Revert "cleanup""

This reverts commit cea33d9a5d.

* Revert "Revert "true dependency""

This reverts commit 4d1b460b

* cleanup

* Forcing version 0.1.5

* Update pyproject.toml

---------

Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-04-20 12:20:13 -03:00
João Moura
f3bfabb149 Adding manager llm 2024-04-16 16:50:44 -03:00
Joseph Bastulli
74f23ada73 fix: swapped the task callback assignment (#443) 2024-04-16 15:54:42 -03:00
Jack Hayter
e307da43c5 Prevent duplicate TokenCalcHandler callbacks on Agent (#475) 2024-04-16 15:54:02 -03:00
Lennart J. Kurzweg
19e0e3fcb4 Added optional manager_agent parameter (#474)
* Added optional manager_agent parameter

* Update crew.py

---------

Co-authored-by: Lennart J. Kurzweg (Nx2) <git@nx2.site>
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-04-16 08:18:36 -03:00
João Moura
69964e005d cutting new version with updated cli template 2024-04-11 11:30:30 -03:00
João Moura
e2ad0efbed adding missing import 2024-04-08 02:08:43 -03:00
João Moura
ed1581dee5 removing unnecessary certificate 2024-04-08 01:39:11 -03:00
João Moura
0c717fb24a fixing long temr memory interpolation 2024-04-07 14:55:35 -03:00
rajib
b5ecb36b50 The suggestions were getting split at character level and not at sentence level (#436)
* fix the issue where the suggestions were split at character level

* Update contextual_memory.py

---------

Co-authored-by: rajib76 <rajib76@yahoo.com>
Co-authored-by: João Moura <joaomdmoura@gmail.com>
2024-04-07 02:57:23 -03:00
shivam singh
8314591523 doc: Add documentation to Task model. (#363) 2024-04-05 19:49:36 -03:00
GabeKoga
289e12de5a Feature: Log files (#423)
* log_file

feature: added a new parameter for crew that creates a txt file to log agent execution

* unit tests and documentation

unit test if file is created but not what is inside the file
2024-04-05 19:44:50 -03:00
GabeKoga
459a404563 purple (#428)
changed from yellow to purple for visibility
2024-04-05 18:25:59 -03:00
João Moura
c47c288138 fixing memory docs 2024-04-05 08:59:54 -03:00
João Moura
f7fc61f043 Increasing default max inter 2024-04-05 08:36:09 -03:00
João Moura
9b37aab61a adding max execution time 2024-04-05 08:31:25 -03:00
João Moura
a0ba9a81b2 not overriding llm callbacks 2024-04-05 08:24:20 -03:00
João Moura
fd12609330 Adding new test for crew memory 2024-04-04 15:29:45 -03:00
João Moura
a1e1113101 Removing memory flag from agent in favor of crew memory 2024-04-04 15:29:45 -03:00
João Moura
c45d1c337c TYPO 2024-04-04 15:29:45 -03:00
ftoppi
ce9c343c0a tasks.py: don't call Converter when model response is valid (#406)
* tasks.py: don't call Converter when model response is valid

Try to convert the task output to the expected Pydantic model before sending it to Converter, maybe the model got it right.
2024-04-04 10:11:46 -03:00
João Moura
3c64626c67 setting fake openai key 2024-04-03 06:56:02 -03:00
João Moura
c730d73c10 force reseting db in care of change in dimensions 2024-04-03 05:52:35 -03:00
João Moura
eb17e192c0 Fixing db path 2024-04-03 05:45:59 -03:00
João Moura
6532370c30 creating db file based on package name 2024-04-03 05:22:20 -03:00
João Moura
1da288de85 preparing new version 2024-04-03 05:04:26 -03:00
João Moura
6129c908b8 updating gitignore 2024-04-03 05:04:00 -03:00
João Moura
35f6156ed4 checking crew before using memory 2024-04-03 05:03:43 -03:00
João Moura
0abfcedda8 updating db storage and dependencies 2024-04-02 13:51:05 -03:00
João Moura
bbd4e58b65 Starting i18n language file support 2024-04-01 10:45:17 -03:00
João Moura
f85bf00409 Adding long term, short term, entity and contextual memory 2024-04-01 10:45:17 -03:00
João Moura
3ef2df75fd using .casefold() instead of lower 2024-04-01 10:45:17 -03:00