Vidit Ostwal 704db1d66f fix(llms): map default Claude Sonnet 4.6 to its 1M context window (#7125) ...

* fix(llms): map default Claude Sonnet 4.6 to its 1M context window

Native AnthropicCompletion fell back to 200k for claude-sonnet-4-6, so the default instance understated the documented limit.

* fix(llms): align Anthropic context windows with active Claude models

Retired Claude 3/2/Instant IDs no longer have dedicated entries; the map now covers the current Claude API lineup and their documented 1M vs 200k windows.

* fix(llms): map Claude Mythos 5 to its documented 1M context window

Native AnthropicCompletion fell back to 200k for claude-mythos-5 even though Anthropic lists a 1M-token window.

2026-08-26 14:09:35 -07:00

..

a2a

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

agents

fix(agents): render message content parts as text, not a python repr (#7109)

2026-08-25 19:34:05 +00:00

cassettes

fix: harden brittle LLM event-bus mocks (#6861)

2026-08-07 19:23:32 +05:30

cli

feat(events): report project creation with the id minted for it (#7074)

2026-08-23 15:59:56 -03:00

config

…

crew

…

events

fix: harden brittle LLM event-bus mocks (#6861)

2026-08-07 19:23:32 +05:30

experimental

feat(skills)!: promote Skills Repository out of experimental (#6579)

2026-07-17 14:21:15 +00:00

hooks

fix: skip interception hooks on crewai-internal flows (#7079)

2026-08-25 09:54:17 -04:00

knowledge

JSON first crews (#6131)

2026-06-14 04:19:48 -03:00

llms

fix(llms): map default Claude Sonnet 4.6 to its 1M context window (#7125)

2026-08-26 14:09:35 -07:00

mcp

[OSS-137] Report MCP HTTP auth failures instead of cancelled connections (#7067)

2026-08-25 16:47:31 +00:00

memory

JSON crew fixes (#6217)

2026-06-18 14:14:54 -03:00

pipeline

…

project

Add repository agents to flow definitions (#6437)

2026-07-02 14:28:01 -07:00

rag

JSON first crews (#6131)

2026-06-14 04:19:48 -03:00

security

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

skills

Lorenze/imp/skills progressive disclosure (#6675)

2026-07-28 09:33:43 -07:00

storage

…

task

…

telemetry

fix(events): record task failures as failures, not as successes (#7073)

2026-08-25 14:30:46 +05:30

tools

fix(telemetry): attribute tool errors to the tool that failed (#7043)

2026-08-18 23:29:19 -03:00

tracing

feat(tracing): record when a trace batch is shared with amp (#6966)

2026-08-12 09:46:11 -07:00

utilities

fix(agents): render message content parts as text, not a python repr (#7109)

2026-08-25 19:34:05 +00:00

__init__.py

…

test_agent_multimodal.py

…

test_async_human_feedback.py

fix(flow): emit the flow lifecycle on a suppressed resume (#7071)

2026-08-21 15:51:20 +05:30

test_callback.py

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

test_checkpoint_cli.py

…

test_checkpoint.py

Validate flow CEL expressions at definition load time (#6224)

2026-06-18 12:18:22 -07:00

test_context.py

…

test_crew_multimodal.py

Fix structured output leaks in tool-calling loops (#5897)

2026-05-27 13:20:53 -07:00

test_crew_thread_safety.py

…

test_crew.py

fix(tools)!: make tool-result caching opt-in instead of on by default (#6509)

2026-07-10 17:37:36 -07:00

test_custom_llm.py

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

test_event_record.py

feat(events): add reset_runtime_state to release accumulated bus state

2026-06-10 16:12:28 -07:00

test_execution_uuid.py

feat(execution): introduce execution context management with UUID sup… (#6988)

2026-08-13 16:29:46 -07:00

test_flow_ask.py

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

test_flow_conversation.py

[OSS-115] Persist custom conversational replies after fallback append (#7026)

2026-08-26 09:41:28 -07:00

test_flow_default_override.py

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

test_flow_definition.py

docs(flow): document declarative conversational flows (#7035)

2026-08-19 20:18:18 +05:30

test_flow_from_definition.py

fix(agents): keep message roles when Agent.kickoff gets a conversation (#7065)

2026-08-24 21:09:36 +00:00

test_flow_human_input_integration.py

Implement message setup and feedback handling in AgentExecutor (#6465)

2026-07-06 15:28:37 -07:00

test_flow_multimodal.py

…

test_flow_persistence_factory.py

feat(storage): pluggable default backends for memory, knowledge, rag, flow (#6079)

2026-06-08 21:14:13 -05:00

test_flow_persistence.py

Remove StateProxy from flow state access (#6327)

2026-06-24 16:37:51 -07:00

test_flow_resumability_regression.py

…

test_flow_usage_metrics.py

fix: aggregate token usage across all LLM calls (#6122)

2026-06-12 12:55:22 -04:00

test_flow_visualization.py

Require explicit CrewAI project definitions (#6358)

2026-06-26 12:07:03 -07:00

test_flow.py

fix(flow): clarify conversational route/handler name collision errors (#6825)

2026-08-06 00:14:55 +05:30

test_guardrail_serialization.py

…

test_hallucination_guardrail.py

…

test_human_feedback_decorator.py

JSON first crews (#6131)

2026-06-14 04:19:48 -03:00

test_human_feedback_integration.py

JSON first crews (#6131)

2026-06-14 04:19:48 -03:00

test_imports.py

…

test_llm_streaming_finish_reason.py

fix: harden brittle LLM event-bus mocks (#6861)

2026-08-07 19:23:32 +05:30

test_llm.py

fix(llms): raise Anthropic default max_tokens so large tool calls survive (#7077)

2026-08-26 19:01:31 +00:00

test_markdown_task.py

…

test_multimodal_validation.py

…

test_project.py

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

test_stream_frames.py

Define stream frame protocol for flows (#6391)

2026-06-30 10:53:48 -07:00

test_streaming_integration.py

Define stream frame protocol for flows (#6391)

2026-06-30 10:53:48 -07:00

test_streaming.py

Define stream frame protocol for flows (#6391)

2026-06-30 10:53:48 -07:00

test_task_guardrails.py

chore(crewai): drop self-explanatory comments

2026-05-26 10:23:33 -07:00

test_task.py

feat(tools): surface tool failures instead of reporting them as success (#6712)

2026-07-29 19:30:14 -07:00

test_tool_cache_default.py

fix(tools)!: make tool-result caching opt-in instead of on by default (#6509)

2026-07-10 17:37:36 -07:00

test_usage_shape_parity.py

fix(output): expose token usage under both names on agent and crew results (#6507)

2026-07-10 20:29:34 -03:00

utils.py

…