Files
crewAI/lib/crewai/tests/llms/bedrock/test_bedrock.py
João Moura bb477f8a91
Some checks failed
CodeQL Advanced / Analyze (actions) (push) Has been cancelled
CodeQL Advanced / Analyze (python) (push) Has been cancelled
Check Documentation Broken Links / Check broken links (push) Has been cancelled
Vulnerability Scan / pip-audit (push) Has been cancelled
Nightly Canary Release / Check for new commits (push) Has been cancelled
Nightly Canary Release / Build nightly packages (push) Has been cancelled
Nightly Canary Release / Publish nightly to PyPI (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
JSON first crews (#6131)
* feat(cli): introduce JSON crew project support and TUI enhancements

- Added support for creating and running JSON-defined crew projects, allowing users to scaffold projects with a new `create_json_crew.py` file.
- Implemented a full-screen Textual TUI for crew execution in `crew_run_tui.py`, enhancing user interaction with a two-column layout.
- Updated `run_crew.py` to prioritize JSON crew projects and added daemon mode for running without TUI.
- Introduced interactive pickers in `tui_picker.py` for improved CLI prompts.
- Enhanced validation for JSON crew files in `validate.py` to ensure proper structure and agent definitions.
- Updated `.gitignore` to exclude demo and crewai directories.

* feat: update LLM model references to gpt-5.4-mini

- Changed default LLM model from gpt-4o-mini to gpt-5.4-mini across various files, including CLI options, JSON crew configurations, and agent definitions.
- Enhanced benchmark and human feedback functionalities to utilize the new model.
- Improved user interface elements in the TUI for better interaction and feedback during execution.
- Added support for new skills directory in JSON crew project creation.

* feat(benchmark): add crew-level benchmarking functionality

- Introduced a new `benchmark` command in the CLI for crew-level benchmarking, allowing users to specify agents, models, and timeout settings.
- Implemented `CrewBenchmarkCase` to handle crew-level benchmark cases with inputs and criteria.
- Enhanced the benchmark runner to support progress tracking and detailed reporting of results for multiple models.
- Added tests for loading crew benchmark cases and validating their structure.
- Updated existing benchmark functions to accommodate the new crew-level execution model.

* feat(cli): enhance JSON crew project functionality and TUI improvements

- Added optional agent-level guardrails and advanced options in JSON crew configurations to improve output validation and flexibility.
- Updated the TUI to better handle plan step statuses, including visual indicators for task completion and failure.
- Introduced methods for parsing and managing step observation events, ensuring accurate updates to task statuses during execution.
- Enhanced validation for JSON crew projects, ensuring proper structure and error handling for agent and task definitions.
- Added comprehensive tests for new features and validation logic, ensuring robustness in JSON crew project handling.

* refactor(cli): streamline JSON crew project handling and improve validation

- Refactored JSON crew project loading and validation logic to enhance clarity and maintainability.
- Introduced utility functions for finding JSON crew files, improving code reuse across modules.
- Removed deprecated benchmark functionality and associated tests to simplify the codebase.
- Updated CLI commands to utilize the new JSON project structure, ensuring compatibility with recent changes.
- Enhanced test coverage for JSON crew project features, ensuring robust validation and error handling.

* feat(cli): enhance activity log navigation and focus management

- Added functionality to focus on the activity log when navigating through log entries.
- Implemented refresh logic for the log panel to ensure updates are displayed correctly during navigation.
- Improved keyboard navigation for log entries, allowing users to expand and scroll through logs seamlessly.
- Added tests to verify the correct behavior of log navigation and focus management in the TUI.

* feat(cli): enhance JSON crew project interaction and input handling

- Introduced a new function to enable prompt line editing for better user experience during input prompts.
- Updated the JSON crew project wizards to show interpolation hints for dynamic values, improving user guidance.
- Enhanced the handling of missing input placeholders by prompting users for required values during crew setup.
- Refactored the crew run logic to ensure proper loading and preparation of JSON-defined crews, including runtime input management.
- Added tests to verify the correct behavior of new input handling features and JSON crew project interactions.

* feat(cli): improve crew project input prompts and event handling

- Enhanced the `_prompt_text` function to allow for configurable spacing before prompts, improving user experience during input collection.
- Updated the wizards for agent and task creation to utilize the new prompt configuration, ensuring a more compact and streamlined interaction.
- Introduced new plan step lifecycle events (`PlanStepStartedEvent`, `PlanStepCompletedEvent`) to better track the execution status of plan steps.
- Refactored the step executor to emit these events during the execution of tasks, improving observability and debugging capabilities.
- Added tests to verify the correct behavior of new prompt handling and event emissions during crew project execution.

* fix: refine json-first crew interactions

* fix: prioritize common json crew tools

* fix: make json crew more tools expandable

* fix: show json crew tools by category

* feat(memory): update default embedder to OpenAI text-embedding-3-large and enhance memory compatibility

- Changed the default embedding model for Memory to OpenAI text-embedding-3-large, which uses 3072-dimensional vectors.
- Added warnings regarding compatibility issues with existing local memory stores created with 1536-dimensional embeddings.
- Updated documentation to reflect the new default embedder and its configuration options.
- Enhanced the CLI and codebase to support the new embedding model across various components, ensuring a seamless transition for users.

* fix: address PR review feedback for JSON-first crews

Review blockers:
- Forward trained_agents_file to JSON crews: crewai run -f now exports
  CREWAI_TRAINED_AGENTS_FILE for the in-process JSON crew path
- Wizard agent picker: Esc/cancel now reprompts instead of silently
  assigning the first agent
- JSON tool resolution hard-fails: unknown tool names, missing custom
  tool files, and invalid custom tool modules raise JSONProjectError
  with actionable messages instead of warn-and-continue
- Embedding dimension mismatch: LanceDB and Qdrant Edge storages raise
  EmbeddingDimensionMismatchError with reset/pin guidance instead of
  silently zero-filling vectors or returning empty search results
- Custom tool code execution documented in loader docstring and the
  scaffolded project README

CI fixes:
- ruff format across lib/
- All 133 PR-introduced mypy errors fixed (llm.py lazy-litellm and
  cli.py lazy command shims now use TYPE_CHECKING imports; textual
  is_mounted misuse fixed; pick_many overloads; misc annotations)

Bot review comments:
- Empty except blocks now have explanatory comments or debug logging
- Removed unused _C_BG/_C_PANEL/_C_BORDER globals and redundant
  import re; tests use a single import style for create_json_crew

Tests: trained-agents propagation, wizard cancel, tool resolution
failures, and dimension mismatch guidance.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix: address second round of PR review comments

Cursor Bugbot:
- Wizard agent slugs: strip to [a-z0-9_] and fall back to agent_<n> so
  symbol-only roles can't produce an empty agents/.jsonc filename
- Wizard task names: dedupe against prior task names and fall back to
  task_<n> for symbol-only descriptions

CodeRabbit:
- Agent.message(): import Task explicitly at runtime instead of relying
  on the namespace injection done by crewai/__init__
- Async executor: move the native-tools-unsupported fallback from
  _ainvoke_loop_react (self-recursion) to _ainvoke_loop_native_tools,
  mirroring the sync implementation
- StepExecutor downgrade: keep the in-step conversation and append the
  text-tooling instructions instead of rebuilding messages, so completed
  native tool calls are not re-executed
- crewai-files: extension-based MIME lookup now runs before byte
  sniffing so csv/xml types are not degraded to text/plain
- Memory storages: validate every record in a save() batch against a
  consistent embedding dimension (LanceDB previously checked only the
  first record); added mixed-batch tests
- _print_post_tui_summary now typed against CrewRunApp
- Docs: Azure OpenAI default embedder change called out in the memory
  migration warning and provider table

Code quality bots:
- Removed unused _C_YELLOW/_C_CYAN (crew_run_tui) and _GREEN (tui_picker)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(cli): accordion tool picker in JSON crew wizard

The flat tool list had grown to ~90 rows. The picker now shows:
- Common tools always visible at the top
- Every other category as a single expandable row with tool and
  selection counts (e.g. "Search & Research  (27 tools, 2 selected)")
- Expanding a category collapses the previously expanded one
- Selections persist across expand/collapse via new preselected
  support in pick_many; cursor follows the toggled category row

tui_picker gains preselected + initial_cursor options on pick_many,
and Esc in multi-select now confirms the current selection instead of
discarding it (required so collapsing can't silently drop choices).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* refactor(cli): remove --daemon flag from crewai run

The flag only affected JSON crew projects — classic and flow projects
ignored it entirely, which made the behavior inconsistent. Removed the
option, the daemon code path (_run_json_crew_daemon), and its helper
(_load_json_crew_with_inputs).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* test: update run command tests after --daemon removal

lib/crewai/tests/cli/test_run_crew.py still asserted the old
run_crew(trained_agents_file=..., daemon=False) call signature.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(cli): exit codes, mid-run quit, async statuses, hyphen placeholders

Addresses the latest Bugbot review round:

- Failed JSON crew runs now exit non-zero (SystemExit(1)) so scripts
  and CI don't treat failures as success, mirroring the classic path
- Quitting the TUI mid-run now ends the process (os._exit(130));
  kickoff runs in a thread worker that cannot be force-cancelled, so
  letting the CLI return would leave LLM/tool work burning tokens in
  the background
- Sidebar task statuses are now async-safe: completion/failure events
  resolve the task's own row via identity instead of assuming the most
  recently started task, and starting a task no longer blanket-marks
  earlier active rows as done
- The runtime-input prompt regex now accepts hyphenated placeholder
  names ({my-topic}), matching kickoff's interpolation pattern

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix: validation safety, custom tool sandboxing, TUI log integrity, memory error surfacing

- Deploy validation no longer executes project code: validation mode
  checks tool declarations structurally (well-formed entries, custom
  tool file exists) without importing or instantiating anything.
  custom:<name> resolution only happens on the actual run path.
- custom:<name> is constrained to [A-Za-z_][A-Za-z0-9_]* and the
  resolved path must stay inside the project's tools/ directory, so
  custom:../foo or absolute-path names cannot execute code outside it.
  Tool paths resolve relative to the crew project root, not cwd.
- TUI task logs are built from per-task state captured at task start
  (idx, description, agent, start time); an out-of-order completion
  takes its output from the event and no longer steals or resets the
  current task's streamed steps/output.
- EmbeddingDimensionMismatchError now inherits ValueError instead of
  RuntimeError so background saves surface it through
  MemorySaveFailedEvent instead of silently dropping the save; the
  shutdown catch in _background_encode_batch is narrowed to the
  "cannot schedule new futures" case.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(cli): declared project type wins over crew.json presence

A flow project that also contains a crew.json(c) file now runs and
validates as the flow it declares in pyproject.toml instead of being
hijacked by the JSON crew path. Both crewai run (_has_json_crew) and
deploy validation (_is_json_crew) check tool.crewai.type; a missing or
unreadable pyproject still means a bare JSON crew project.

Also documents why StepObservationFailedEvent intentionally marks the
plan step "done": the event signals an observer failure, not a step
failure, and the executor continues past it.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(cli): type the declared_type locals so mypy stays clean

Comparing an Any-typed .get() chain returns Any, which tripped
no-any-return on the previous commit.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-06-14 04:19:48 -03:00

1189 lines
39 KiB
Python

import os
import sys
import types
from unittest.mock import patch, MagicMock
import pytest
from crewai.llm import LLM
from crewai.crew import Crew
from crewai.agent import Agent
from crewai.task import Task
def _create_bedrock_mocks():
"""Helper to create Bedrock mocks."""
mock_session_class = MagicMock()
mock_session_instance = MagicMock()
mock_client = MagicMock()
# Set up default mock responses to prevent hanging
default_response = {
'output': {
'message': {
'role': 'assistant',
'content': [
{'text': 'Test response'}
]
}
},
'usage': {
'inputTokens': 10,
'outputTokens': 5,
'totalTokens': 15
}
}
mock_client.converse.return_value = default_response
mock_client.converse_stream.return_value = {'stream': []}
mock_session_instance.client.return_value = mock_client
mock_session_class.return_value = mock_session_instance
return mock_session_class, mock_client
@pytest.fixture(autouse=True)
def mock_aws_credentials():
"""Mock AWS credentials and boto3 Session for tests only if real credentials are not set."""
if "AWS_ACCESS_KEY_ID" in os.environ and "AWS_SECRET_ACCESS_KEY" in os.environ:
yield None, None
return
with patch.dict(os.environ, {
"AWS_ACCESS_KEY_ID": "test-access-key",
"AWS_SECRET_ACCESS_KEY": "test-secret-key",
"AWS_DEFAULT_REGION": "us-east-1"
}):
# Mock boto3 Session to prevent actual AWS connections
with patch('crewai.llms.providers.bedrock.completion.Session') as mock_session_class:
mock_session_instance = MagicMock()
mock_client = MagicMock()
# Set up default mock responses to prevent hanging
default_response = {
'output': {
'message': {
'role': 'assistant',
'content': [
{'text': 'Test response'}
]
}
},
'usage': {
'inputTokens': 10,
'outputTokens': 5,
'totalTokens': 15
}
}
mock_client.converse.return_value = default_response
mock_client.converse_stream.return_value = {'stream': []}
mock_session_instance.client.return_value = mock_client
mock_session_class.return_value = mock_session_instance
yield mock_session_class, mock_client
@pytest.fixture
def bedrock_mocks():
"""Fixture that always provides Bedrock mocks, regardless of real credentials.
Use this fixture for tests that explicitly need to test mock behavior.
"""
with patch.dict(os.environ, {
"AWS_ACCESS_KEY_ID": "test-access-key",
"AWS_SECRET_ACCESS_KEY": "test-secret-key",
"AWS_DEFAULT_REGION": "us-east-1"
}):
with patch('crewai.llms.providers.bedrock.completion.Session') as mock_session_class:
mock_session_instance = MagicMock()
mock_client = MagicMock()
default_response = {
'output': {
'message': {
'role': 'assistant',
'content': [
{'text': 'Test response'}
]
}
},
'usage': {
'inputTokens': 10,
'outputTokens': 5,
'totalTokens': 15
}
}
mock_client.converse.return_value = default_response
mock_client.converse_stream.return_value = {'stream': []}
mock_session_instance.client.return_value = mock_client
mock_session_class.return_value = mock_session_instance
yield mock_session_class, mock_client
def test_bedrock_completion_is_used_when_bedrock_provider():
"""
Test that BedrockCompletion from completion.py is used when LLM uses provider 'bedrock'
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
assert llm.__class__.__name__ == "BedrockCompletion"
assert llm.provider == "bedrock"
assert llm.model == "anthropic.claude-3-5-sonnet-20241022-v2:0"
def test_bedrock_completion_module_is_imported():
"""
Test that the completion module is properly imported when using Bedrock provider
"""
module_name = "crewai.llms.providers.bedrock.completion"
if module_name in sys.modules:
del sys.modules[module_name]
LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
assert module_name in sys.modules
completion_mod = sys.modules[module_name]
assert isinstance(completion_mod, types.ModuleType)
assert hasattr(completion_mod, 'BedrockCompletion')
def test_native_bedrock_raises_error_when_initialization_fails():
"""
Test that LLM raises ImportError when native Bedrock completion fails.
With the new behavior, when a native provider is in SUPPORTED_NATIVE_PROVIDERS
but fails to instantiate, we raise an ImportError instead of silently falling back.
This provides clearer error messages to users about missing dependencies.
"""
with patch('crewai.llm.LLM._get_native_provider') as mock_get_provider:
class FailingCompletion:
def __init__(self, *args, **kwargs):
raise Exception("Native AWS Bedrock SDK failed")
mock_get_provider.return_value = FailingCompletion
with pytest.raises(ImportError) as excinfo:
LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
assert "Error importing native provider" in str(excinfo.value)
assert "Native AWS Bedrock SDK failed" in str(excinfo.value)
def test_bedrock_completion_initialization_parameters():
"""
Test that BedrockCompletion is initialized with correct parameters
"""
llm = LLM(
model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0",
temperature=0.7,
max_tokens=2000,
top_p=0.9,
top_k=40,
region_name="us-west-2"
)
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm, BedrockCompletion)
assert llm.model == "anthropic.claude-3-5-sonnet-20241022-v2:0"
assert llm.temperature == 0.7
assert llm.max_tokens == 2000
assert llm.top_p == 0.9
assert llm.top_k == 40
assert llm.region_name == "us-west-2"
def test_bedrock_specific_parameters():
"""
Test Bedrock-specific parameters like stop_sequences and streaming
"""
llm = LLM(
model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0",
stop_sequences=["Human:", "Assistant:"],
stream=True,
region_name="us-east-1"
)
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm, BedrockCompletion)
assert llm.stop_sequences == ["Human:", "Assistant:"]
assert llm.stream == True
assert llm.region_name == "us-east-1"
def test_bedrock_completion_call():
"""
Test that BedrockCompletion call method works
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(llm, 'call', return_value="Hello! I'm Claude on Bedrock, ready to help.") as mock_call:
result = llm.call("Hello, how are you?")
assert result == "Hello! I'm Claude on Bedrock, ready to help."
mock_call.assert_called_once_with("Hello, how are you?")
def test_bedrock_completion_called_during_crew_execution():
"""
Test that BedrockCompletion.call is actually invoked when running a crew
"""
bedrock_llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(bedrock_llm, 'call', return_value="Tokyo has 14 million people.") as mock_call:
agent = Agent(
role="Research Assistant",
goal="Find population info",
backstory="You research populations.",
llm=bedrock_llm,
)
task = Task(
description="Find Tokyo population",
expected_output="Population number",
agent=agent,
)
crew = Crew(agents=[agent], tasks=[task])
result = crew.kickoff()
assert mock_call.called
assert "14 million" in str(result)
@pytest.mark.skip(reason="Crew execution test - may hang, needs investigation")
def test_bedrock_completion_call_arguments():
"""
Test that BedrockCompletion.call is invoked with correct arguments
"""
bedrock_llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(bedrock_llm, 'call') as mock_call:
mock_call.return_value = "Task completed successfully."
agent = Agent(
role="Test Agent",
goal="Complete a simple task",
backstory="You are a test agent.",
llm=bedrock_llm
)
task = Task(
description="Say hello world",
expected_output="Hello world",
agent=agent,
)
crew = Crew(agents=[agent], tasks=[task])
crew.kickoff()
assert mock_call.called
call_args = mock_call.call_args
assert call_args is not None
messages = call_args[0][0]
assert isinstance(messages, (str, list))
if isinstance(messages, str):
assert "hello world" in messages.lower()
elif isinstance(messages, list):
message_content = str(messages).lower()
assert "hello world" in message_content
def test_multiple_bedrock_calls_in_crew():
"""
Test that BedrockCompletion.call is invoked multiple times for multiple tasks
"""
bedrock_llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(bedrock_llm, 'call') as mock_call:
mock_call.return_value = "Task completed."
agent = Agent(
role="Multi-task Agent",
goal="Complete multiple tasks",
backstory="You can handle multiple tasks.",
llm=bedrock_llm
)
task1 = Task(
description="First task",
expected_output="First result",
agent=agent,
)
task2 = Task(
description="Second task",
expected_output="Second result",
agent=agent,
)
crew = Crew(
agents=[agent],
tasks=[task1, task2]
)
crew.kickoff()
assert mock_call.call_count >= 2 # At least one call per task
for call in mock_call.call_args_list:
assert len(call[0]) > 0
messages = call[0][0]
assert messages is not None
def test_bedrock_completion_with_tools():
"""
Test that BedrockCompletion.call is invoked with tools when agent has tools
"""
from crewai.tools import tool
@tool
def sample_tool(query: str) -> str:
"""A sample tool for testing"""
return f"Tool result for: {query}"
bedrock_llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(bedrock_llm, 'call') as mock_call:
mock_call.return_value = "Task completed with tools."
agent = Agent(
role="Tool User",
goal="Use tools to complete tasks",
backstory="You can use tools.",
llm=bedrock_llm,
tools=[sample_tool]
)
task = Task(
description="Use the sample tool",
expected_output="Tool usage result",
agent=agent,
)
crew = Crew(agents=[agent], tasks=[task])
crew.kickoff()
assert mock_call.called
call_args = mock_call.call_args
call_kwargs = call_args[1] if len(call_args) > 1 else {}
if 'tools' in call_kwargs:
assert call_kwargs['tools'] is not None
assert len(call_kwargs['tools']) > 0
@pytest.mark.timeout(180)
def test_bedrock_raises_error_when_model_not_found(bedrock_mocks):
"""Test that BedrockCompletion raises appropriate error when model not found"""
from botocore.exceptions import ClientError
_, mock_client = bedrock_mocks
error_response = {
'Error': {
'Code': 'ResourceNotFoundException',
'Message': 'Could not resolve the foundation model from the model identifier'
}
}
mock_client.converse.side_effect = ClientError(error_response, 'converse')
llm = LLM(model="bedrock/model-doesnt-exist")
with pytest.raises(Exception): # Should raise some error for unsupported model
llm.call("Hello")
def test_bedrock_aws_credentials_configuration():
"""
Test that AWS credentials configuration works properly
"""
aws_access_key_id = "test-access-key"
aws_secret_access_key = "test-secret-key"
aws_region_name = "us-east-1"
with patch.dict(os.environ, {
"AWS_ACCESS_KEY_ID": aws_access_key_id,
"AWS_SECRET_ACCESS_KEY": aws_secret_access_key,
"AWS_DEFAULT_REGION": aws_region_name
}):
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm, BedrockCompletion)
assert llm.region_name == aws_region_name
assert llm.aws_access_key_id == aws_access_key_id
assert llm.aws_secret_access_key == aws_secret_access_key
# Test with litellm environment variables
with patch.dict(os.environ, {
"AWS_ACCESS_KEY_ID": aws_access_key_id,
"AWS_SECRET_ACCESS_KEY": aws_secret_access_key,
"AWS_REGION_NAME": aws_region_name
}):
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm, BedrockCompletion)
assert llm.region_name == aws_region_name
llm_explicit = LLM(
model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0",
aws_access_key_id="explicit-key",
aws_secret_access_key="explicit-secret",
region_name="us-west-2"
)
assert isinstance(llm_explicit, BedrockCompletion)
assert llm_explicit.region_name == "us-west-2"
def test_bedrock_model_capabilities():
"""
Test that model capabilities are correctly identified
"""
llm_claude = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm_claude, BedrockCompletion)
assert llm_claude.is_claude_model == True
assert llm_claude.supports_tools == True
# Test other Bedrock model
llm_titan = LLM(model="bedrock/amazon.titan-text-express-v1")
assert isinstance(llm_titan, BedrockCompletion)
assert llm_titan.supports_tools == True
def test_bedrock_inference_config():
"""
Test that inference config is properly prepared
"""
llm = LLM(
model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0",
temperature=0.7,
top_p=0.9,
top_k=40,
max_tokens=1000
)
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm, BedrockCompletion)
config = llm._get_inference_config()
assert 'temperature' in config
assert config['temperature'] == 0.7
assert 'topP' in config
assert config['topP'] == 0.9
assert 'maxTokens' in config
assert config['maxTokens'] == 1000
assert 'topK' in config
assert config['topK'] == 40
def test_bedrock_model_detection():
"""
Test that various Bedrock model formats are properly detected
"""
# Test Bedrock model naming patterns
bedrock_test_cases = [
"bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0",
"bedrock/anthropic.claude-3-haiku-20240307-v1:0",
"bedrock/amazon.titan-text-express-v1",
"bedrock/meta.llama3-70b-instruct-v1:0"
]
for model_name in bedrock_test_cases:
llm = LLM(model=model_name)
from crewai.llms.providers.bedrock.completion import BedrockCompletion
assert isinstance(llm, BedrockCompletion), f"Failed for model: {model_name}"
def test_bedrock_supports_stop_words():
"""
Test that Bedrock models support stop sequences
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
assert llm.supports_stop_words() == True
def test_bedrock_context_window_size():
"""
Test that Bedrock models return correct context window sizes
"""
llm_claude = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
context_size_claude = llm_claude.get_context_window_size()
assert context_size_claude > 150000 # Should be substantial (200K tokens with ratio)
llm_titan = LLM(model="bedrock/amazon.titan-text-express-v1")
context_size_titan = llm_titan.get_context_window_size()
assert context_size_titan > 5000
def test_bedrock_message_formatting():
"""
Test that messages are properly formatted for Bedrock Converse API
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
test_messages = [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Hello"},
{"role": "assistant", "content": "Hi there!"},
{"role": "user", "content": "How are you?"}
]
formatted_messages, system_message = llm._format_messages_for_converse(test_messages)
# System message should be extracted
assert system_message == "You are a helpful assistant."
# Remaining messages should be in Converse format
assert len(formatted_messages) >= 3
assert formatted_messages[0]["role"] == "user"
assert formatted_messages[1]["role"] == "assistant"
# Messages should have content array with text
assert isinstance(formatted_messages[0]["content"], list)
assert "text" in formatted_messages[0]["content"][0]
def test_bedrock_streaming_parameter():
"""
Test that streaming parameter is properly handled
"""
llm_no_stream = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0", stream=False)
assert llm_no_stream.stream == False
llm_stream = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0", stream=True)
assert llm_stream.stream == True
def test_bedrock_tool_conversion():
"""
Test that tools are properly converted to Bedrock Converse format
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
crewai_tools = [{
"type": "function",
"function": {
"name": "test_tool",
"description": "A test tool",
"parameters": {
"type": "object",
"properties": {
"query": {"type": "string", "description": "Search query"}
},
"required": ["query"]
}
}
}]
bedrock_tools = llm._format_tools_for_converse(crewai_tools)
assert len(bedrock_tools) == 1
# Bedrock tools should have toolSpec structure
assert "toolSpec" in bedrock_tools[0]
assert bedrock_tools[0]["toolSpec"]["name"] == "test_tool"
assert bedrock_tools[0]["toolSpec"]["description"] == "A test tool"
assert "inputSchema" in bedrock_tools[0]["toolSpec"]
def test_bedrock_environment_variable_credentials(bedrock_mocks):
"""
Test that AWS credentials are properly loaded from environment
"""
mock_session_class, _ = bedrock_mocks
mock_session_class.reset_mock()
with patch.dict(os.environ, {
"AWS_ACCESS_KEY_ID": "test-access-key-123",
"AWS_SECRET_ACCESS_KEY": "test-secret-key-456"
}):
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
assert mock_session_class.called
call_kwargs = mock_session_class.call_args[1] if mock_session_class.call_args else {}
assert call_kwargs.get('aws_access_key_id') == "test-access-key-123"
assert call_kwargs.get('aws_secret_access_key') == "test-secret-key-456"
def test_bedrock_token_usage_tracking():
"""
Test that token usage is properly tracked for Bedrock responses
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
# Mock the Bedrock response with usage information
with patch.object(llm._client, 'converse') as mock_converse:
mock_response = {
'output': {
'message': {
'role': 'assistant',
'content': [
{'text': 'test response'}
]
}
},
'usage': {
'inputTokens': 50,
'outputTokens': 25,
'totalTokens': 75
}
}
mock_converse.return_value = mock_response
result = llm.call("Hello")
assert result == "test response"
assert llm._token_usage['prompt_tokens'] == 50
assert llm._token_usage['completion_tokens'] == 25
assert llm._token_usage['total_tokens'] == 75
def test_bedrock_tool_use_conversation_flow():
"""
Test that the Bedrock completion properly handles tool use conversation flow
"""
from unittest.mock import Mock
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
def mock_weather_tool(location: str) -> str:
return f"The weather in {location} is sunny and 75°F"
available_functions = {"get_weather": mock_weather_tool}
# Mock the Bedrock client responses
with patch.object(llm._client, 'converse') as mock_converse:
tool_use_response = {
'output': {
'message': {
'role': 'assistant',
'content': [
{
'toolUse': {
'toolUseId': 'tool-123',
'name': 'get_weather',
'input': {'location': 'San Francisco'}
}
}
]
}
},
'usage': {
'inputTokens': 100,
'outputTokens': 50,
'totalTokens': 150
}
}
final_response = {
'output': {
'message': {
'role': 'assistant',
'content': [
{'text': 'Based on the weather data, it is sunny and 75°F in San Francisco.'}
]
}
},
'usage': {
'inputTokens': 120,
'outputTokens': 30,
'totalTokens': 150
}
}
mock_converse.side_effect = [tool_use_response, final_response]
messages = [{"role": "user", "content": "What's the weather like in San Francisco?"}]
result = llm.call(
messages=messages,
available_functions=available_functions
)
assert "sunny" in result.lower() or "75" in result
# Verify that the API was called twice (once for tool use, once for final answer)
assert mock_converse.call_count == 2
def test_bedrock_handles_cohere_conversation_requirements():
"""
Test that Bedrock properly handles Cohere model's requirement for user message at end
"""
llm = LLM(model="bedrock/cohere.command-r-plus-v1:0")
test_messages = [
{"role": "user", "content": "Hello"},
{"role": "assistant", "content": "Hi there!"}
]
formatted_messages, system_message = llm._format_messages_for_converse(test_messages)
# For Cohere models, should add a user message at the end
assert formatted_messages[-1]["role"] == "user"
assert "continue" in formatted_messages[-1]["content"][0]["text"].lower()
def test_bedrock_client_error_handling():
"""
Test that Bedrock properly handles various AWS client errors
"""
from botocore.exceptions import ClientError
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(llm._client, 'converse') as mock_converse:
error_response = {
'Error': {
'Code': 'ValidationException',
'Message': 'Invalid request format'
}
}
mock_converse.side_effect = ClientError(error_response, 'converse')
with pytest.raises(ValueError) as exc_info:
llm.call("Hello")
assert "validation" in str(exc_info.value).lower()
with patch.object(llm._client, 'converse') as mock_converse:
error_response = {
'Error': {
'Code': 'ThrottlingException',
'Message': 'Rate limit exceeded'
}
}
mock_converse.side_effect = ClientError(error_response, 'converse')
with pytest.raises(RuntimeError) as exc_info:
llm.call("Hello")
assert "throttled" in str(exc_info.value).lower()
def test_bedrock_stop_sequences_sync():
"""Test that stop and stop_sequences attributes stay synchronized."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
llm.stop = ["\nObservation:", "\nThought:"]
assert list(llm.stop_sequences) == ["\nObservation:", "\nThought:"]
assert llm.stop == ["\nObservation:", "\nThought:"]
llm.stop = "\nFinal Answer:"
assert list(llm.stop_sequences) == ["\nFinal Answer:"]
assert llm.stop == ["\nFinal Answer:"]
llm.stop = None
assert list(llm.stop_sequences) == []
assert llm.stop == []
def test_bedrock_stop_sequences_sent_to_api():
"""Test that stop_sequences are properly sent to the Bedrock API."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
# Set stop sequences via the stop attribute (simulating CrewAgentExecutor)
llm.stop = ["\nObservation:", "\nThought:"]
with patch.object(llm._client, 'converse') as mock_converse:
mock_response = {
'output': {
'message': {
'role': 'assistant',
'content': [{'text': 'Hello'}]
}
},
'usage': {
'inputTokens': 10,
'outputTokens': 5,
'totalTokens': 15
}
}
mock_converse.return_value = mock_response
llm.call("Say hello in one word")
call_kwargs = mock_converse.call_args[1]
assert "inferenceConfig" in call_kwargs
assert "stopSequences" in call_kwargs["inferenceConfig"]
assert call_kwargs["inferenceConfig"]["stopSequences"] == ["\nObservation:", "\nThought:"]
# Agent Kickoff Structured Output Tests
@pytest.mark.vcr()
def test_bedrock_agent_kickoff_structured_output_without_tools():
"""
Test that agent kickoff returns structured output without tools.
This tests native structured output handling for Bedrock models.
"""
from pydantic import BaseModel, Field
class AnalysisResult(BaseModel):
"""Structured output for analysis results."""
topic: str = Field(description="The topic analyzed")
key_points: list[str] = Field(description="Key insights from the analysis")
summary: str = Field(description="Brief summary of findings")
agent = Agent(
role="Analyst",
goal="Provide structured analysis on topics",
backstory="You are an expert analyst who provides clear, structured insights.",
llm=LLM(model="bedrock/us.anthropic.claude-sonnet-4-6"),
tools=[],
verbose=True,
)
result = agent.kickoff(
messages="Analyze the benefits of remote work briefly. Keep it concise.",
response_format=AnalysisResult,
)
assert result.pydantic is not None, "Expected pydantic output but got None"
assert isinstance(result.pydantic, AnalysisResult), f"Expected AnalysisResult but got {type(result.pydantic)}"
assert result.pydantic.topic, "Topic should not be empty"
assert len(result.pydantic.key_points) > 0, "Should have at least one key point"
assert result.pydantic.summary, "Summary should not be empty"
@pytest.mark.vcr()
def test_bedrock_agent_kickoff_structured_output_with_tools():
"""
Test that agent kickoff returns structured output after using tools.
This tests post-tool-call structured output handling for Bedrock models.
"""
from pydantic import BaseModel, Field
from crewai.tools import tool
class CalculationResult(BaseModel):
"""Structured output for calculation results."""
operation: str = Field(description="The mathematical operation performed")
result: int = Field(description="The result of the calculation")
explanation: str = Field(description="Brief explanation of the calculation")
@tool
def add_numbers(a: int, b: int) -> int:
"""Add two numbers together and return the sum."""
return a + b
agent = Agent(
role="Calculator",
goal="Perform calculations using available tools",
backstory="You are a calculator assistant that uses tools to compute results.",
llm=LLM(model="bedrock/us.anthropic.claude-sonnet-4-6"),
tools=[add_numbers],
verbose=True,
)
result = agent.kickoff(
messages="Calculate 15 + 27 using your add_numbers tool. Report the result.",
response_format=CalculationResult,
)
assert result.pydantic is not None, "Expected pydantic output but got None"
assert isinstance(result.pydantic, CalculationResult), f"Expected CalculationResult but got {type(result.pydantic)}"
assert result.pydantic.result == 42, f"Expected result 42 but got {result.pydantic.result}"
assert result.pydantic.operation, "Operation should not be empty"
assert result.pydantic.explanation, "Explanation should not be empty"
def test_bedrock_groups_three_tool_results():
"""Consecutive tool results should be grouped into one Bedrock user message."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
messages = [
{"role": "user", "content": "Use all three tools, then continue."},
{
"role": "assistant",
"content": "",
"tool_calls": [
{
"id": "tool-1",
"type": "function",
"function": {
"name": "lookup_weather",
"arguments": '{"location": "New York"}',
},
},
{
"id": "tool-2",
"type": "function",
"function": {
"name": "lookup_news",
"arguments": '{"topic": "AI"}',
},
},
{
"id": "tool-3",
"type": "function",
"function": {
"name": "lookup_stock",
"arguments": '{"ticker": "AMZN"}',
},
},
],
},
{"role": "tool", "tool_call_id": "tool-1", "content": "72F and sunny"},
{"role": "tool", "tool_call_id": "tool-2", "content": "AI news summary"},
{"role": "tool", "tool_call_id": "tool-3", "content": "AMZN up 1.2%"},
]
formatted_messages, system_message = llm._format_messages_for_converse(messages)
assert system_message is None
assert [message["role"] for message in formatted_messages] == [
"user",
"assistant",
"user",
]
assert len(formatted_messages[1]["content"]) == 3
tool_results = formatted_messages[2]["content"]
assert len(tool_results) == 3
assert [block["toolResult"]["toolUseId"] for block in tool_results] == [
"tool-1",
"tool-2",
"tool-3",
]
assert [block["toolResult"]["content"][0]["text"] for block in tool_results] == [
"72F and sunny",
"AI news summary",
"AMZN up 1.2%",
]
def test_bedrock_parallel_tool_results_grouped():
"""Regression test for issue #4749.
When an assistant message contains multiple parallel tool calls,
Bedrock requires all corresponding tool results to be grouped
in a single user message. Previously each tool result was emitted
as a separate user message, causing:
ValidationException: Expected toolResult blocks at messages.2.content
"""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
messages = [
{"role": "user", "content": "Calculate 25 + 17 AND 10 * 5"},
{
"role": "assistant",
"content": "",
"tool_calls": [
{
"id": "call_add",
"type": "function",
"function": {"name": "add_tool", "arguments": '{"a": 25, "b": 17}'},
},
{
"id": "call_mul",
"type": "function",
"function": {"name": "multiply_tool", "arguments": '{"a": 10, "b": 5}'},
},
],
},
{"role": "tool", "tool_call_id": "call_add", "content": "42"},
{"role": "tool", "tool_call_id": "call_mul", "content": "50"},
]
converse_msgs, system_msg = llm._format_messages_for_converse(messages)
tool_result_messages = [
m for m in converse_msgs
if m.get("role") == "user"
and any("toolResult" in b for b in m.get("content", []))
]
# There must be exactly ONE user message with tool results (not two)
assert len(tool_result_messages) == 1, (
f"Expected 1 grouped tool-result message, got {len(tool_result_messages)}. "
"Bedrock requires all parallel tool results in a single user message."
)
# That single message must contain both tool results
tool_results = tool_result_messages[0]["content"]
assert len(tool_results) == 2, (
f"Expected 2 toolResult blocks in grouped message, got {len(tool_results)}"
)
tool_use_ids = {
block["toolResult"]["toolUseId"] for block in tool_results
}
assert tool_use_ids == {"call_add", "call_mul"}
def test_bedrock_single_tool_result_still_works():
"""Ensure single tool call still produces a single-block user message."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
messages = [
{"role": "user", "content": "Add 1 + 2"},
{
"role": "assistant",
"content": "",
"tool_calls": [
{
"id": "call_single",
"type": "function",
"function": {"name": "add_tool", "arguments": '{"a": 1, "b": 2}'},
},
],
},
{"role": "tool", "tool_call_id": "call_single", "content": "3"},
]
converse_msgs, _ = llm._format_messages_for_converse(messages)
tool_result_messages = [
m for m in converse_msgs
if m.get("role") == "user"
and any("toolResult" in b for b in m.get("content", []))
]
assert len(tool_result_messages) == 1
assert len(tool_result_messages[0]["content"]) == 1
assert tool_result_messages[0]["content"][0]["toolResult"]["toolUseId"] == "call_single"
def test_bedrock_tool_results_not_merged_across_assistant_messages():
"""Tool results from different assistant turns must NOT be merged."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
messages = [
{"role": "user", "content": "First task"},
{
"role": "assistant",
"content": "",
"tool_calls": [
{
"id": "call_a",
"type": "function",
"function": {"name": "tool_a", "arguments": "{}"},
},
],
},
{"role": "tool", "tool_call_id": "call_a", "content": "result_a"},
{"role": "assistant", "content": "Now doing second task"},
{"role": "user", "content": "Second task"},
{
"role": "assistant",
"content": "",
"tool_calls": [
{
"id": "call_b",
"type": "function",
"function": {"name": "tool_b", "arguments": "{}"},
},
],
},
{"role": "tool", "tool_call_id": "call_b", "content": "result_b"},
]
converse_msgs, _ = llm._format_messages_for_converse(messages)
tool_result_messages = [
m for m in converse_msgs
if m.get("role") == "user"
and any("toolResult" in b for b in m.get("content", []))
]
# Two separate tool-result messages (one per assistant turn)
assert len(tool_result_messages) == 2, (
"Tool results from different assistant turns must remain separate"
)
assert tool_result_messages[0]["content"][0]["toolResult"]["toolUseId"] == "call_a"
assert tool_result_messages[1]["content"][0]["toolResult"]["toolUseId"] == "call_b"
def test_bedrock_cached_token_tracking():
"""Test that cached tokens (cacheReadInputTokenCount) are tracked for Bedrock."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(llm._client, 'converse') as mock_converse:
mock_response = {
'output': {
'message': {
'role': 'assistant',
'content': [{'text': 'test response'}]
}
},
'usage': {
'inputTokens': 100,
'outputTokens': 50,
'totalTokens': 150,
'cacheReadInputTokenCount': 30,
}
}
mock_converse.return_value = mock_response
result = llm.call("Hello")
assert result == "test response"
assert llm._token_usage['prompt_tokens'] == 100
assert llm._token_usage['completion_tokens'] == 50
assert llm._token_usage['total_tokens'] == 150
assert llm._token_usage['cached_prompt_tokens'] == 30
def test_bedrock_cached_token_alternate_key():
"""Test that the alternate key cacheReadInputTokens also works."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(llm._client, 'converse') as mock_converse:
mock_response = {
'output': {
'message': {
'role': 'assistant',
'content': [{'text': 'test response'}]
}
},
'usage': {
'inputTokens': 80,
'outputTokens': 40,
'totalTokens': 120,
'cacheReadInputTokens': 25,
}
}
mock_converse.return_value = mock_response
llm.call("Hello")
assert llm._token_usage['cached_prompt_tokens'] == 25
def test_bedrock_no_cache_tokens_defaults_to_zero():
"""Test that missing cache token keys default to zero."""
llm = LLM(model="bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0")
with patch.object(llm._client, 'converse') as mock_converse:
mock_response = {
'output': {
'message': {
'role': 'assistant',
'content': [{'text': 'test response'}]
}
},
'usage': {
'inputTokens': 60,
'outputTokens': 30,
'totalTokens': 90,
}
}
mock_converse.return_value = mock_response
llm.call("Hello")
assert llm._token_usage['cached_prompt_tokens'] == 0