3 Releases published by 2 people

• 0.117.0

  published Apr 28, 2025

• 0.117.1

  published Apr 28, 2025

• 0.118.0

  published Apr 30, 2025

15 Pull requests merged by 7 people

• Identify parent_flow of Crew and LiteAgent

  #2723 merged May 2, 2025

• Fix crewai reset-memories when Embedding dimension mismatch

  #2737 merged May 2, 2025

• pytest improvements to handle flaky test

  #2726 merged May 1, 2025

• prepping new version

  #2733 merged Apr 30, 2025

• fix: renaming TaskGuardrail to LLMGuardrail

  #2731 merged Apr 30, 2025

• docs: update enterprise docs and change YouTube video embed

  #2728 merged Apr 30, 2025

• Supporting no-code Guardrail creation

  #2636 merged Apr 30, 2025

• docs: remove CrewStructuredTool from public documentation

  #2707 merged Apr 30, 2025

• build: downgrade litellm to 1.167.1

  #2711 merged Apr 30, 2025

• chore: add missing __init__.py files

  #2719 merged Apr 29, 2025

• Remove logging setting from global context

  #2720 merged Apr 29, 2025

• Fixes missing prompt template or system template

  #2408 merged Apr 28, 2025

• Fix Mem0 OSS

  #2604 merged Apr 28, 2025

• upgrade liteLLM to latest version

  #2684 merged Apr 28, 2025

• build: upgrade crewai-tools

  #2705 merged Apr 28, 2025

15 Pull requests opened by 8 people

• Fix #2698: Implement MCP SSE server connection for tools

  #2699 opened Apr 27, 2025

• fix: update OpenInference SDK link for Arize Phoenix integration

  #2703 opened Apr 28, 2025

• Changed chroma db version

  #2704 opened Apr 28, 2025

• test: add windows & macos on tests runner

  #2710 opened Apr 28, 2025

• Fix typos in documentation and configuration files

  #2712 opened Apr 29, 2025

• Added Advance Configuration Docs for Rag Tool

  #2713 opened Apr 29, 2025

• Add task decomposition feature (Issue #2717)

  #2718 opened Apr 29, 2025

• Support list of YAML files for agents_config and tasks_config in CrewBase

  #2722 opened Apr 29, 2025

• Fix issue #2724: Allow specifying trained data file for run command

  #2725 opened Apr 29, 2025

• feat: implement knowledge retrieval events in Agent

  #2727 opened Apr 30, 2025

• Added clarification for contributors who want to run tests

  #2736 opened May 1, 2025

• Fix #2740: Prevent IndexError in ollama_pt() with empty messages

  #2741 opened May 2, 2025

• feat: support to remove `stop` parameter from LLM call

  #2742 opened May 2, 2025

• log agent specific llm assignement

  #2743 opened May 2, 2025

• Fix IndexError in litellm's ollama_pt function when using Ollama/Qwen models with tools

  #2745 opened May 3, 2025

24 Issues closed by 3 people

• [BUG] - Failed to build `chroma-hnswlib==0.7.6` during installation

  #2694 closed May 2, 2025

• [BUG] IndexError: list index out of range in ollama_pt() when messages list is empty or malformed

  #2740 closed May 2, 2025

• [BUG]agent.py:set_knowledge not support CHN

  #2454 closed May 2, 2025

• [BUG] CrewStructuredTool is not usable

  #2700 closed May 2, 2025

• [BUG] Watsonx as embedder is not working - script errors and stops

  #1790 closed May 1, 2025

• [BUG] Openlit with Opentelemetry Version Issues

  #2623 closed May 1, 2025

• [BUG] tool input parameter not passed correctly

  #2288 closed Apr 30, 2025

• [BUG] UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 1980

  #2708 closed Apr 30, 2025

• Can I use a Custom embedder in the crewai_tools?

  #718 closed Apr 29, 2025

• [BUG]Mem0 contextual memory fails

  #2591 closed Apr 29, 2025

• [BUG] Mem0 memory search fails

  #2589 closed Apr 29, 2025

• [BUG] UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 1980: character maps to <undefined>

  #2715 closed Apr 29, 2025

• [BUG] respect_context_window doesn't work

  #2659 closed Apr 28, 2025

• [BUG] FirecrawlApp.crawl_url() takes 2 positional arguments but 3 were given

  #2697 closed Apr 28, 2025

• [BUG] Mem0 (local config) - Memory.search() got an unexpected keyword argument 'metadata'

  #2599 closed Apr 28, 2025

• [FEATURE] Integration with Elastic Search for RAG

  #2671 closed Apr 28, 2025

• Error when connecting mcp sse server tools

  #2698 closed Apr 28, 2025

• [BUG] Expected `thinking` or `redacted_thinking`, but found `text` when using Claude Sonnet 3.7 Thinking

  #2323 closed Apr 28, 2025

• [BUG] Bump litellm to latest version

  #2521 closed Apr 28, 2025

• [BUG] Model's maximum context length causes a crew to throw a python error and stop when scrapping a web page

  #2696 closed Apr 28, 2025

• [BUG] ImportError with MYSQL

  #2361 closed Apr 28, 2025

• [BUG] Version 0.114.0 broke compatibiliity with Python 3.10

  #2575 closed Apr 28, 2025

• [BUG] GithubSearchTool validation errors

  #2101 closed Apr 27, 2025

• [BUG]AttributeError: _ARRAY_API not found ImportError: numpy.core.multiarray failed to import

  #2431 closed Apr 27, 2025

8 Issues opened by 8 people

• [BUG] IndexError in litellm when CrewAI Agent with Tools uses Ollama/Qwen

  #2744 opened May 3, 2025

• [BUG] o3 model support (not o3-mini)

  #2738 opened May 1, 2025

• [FEATURE] Support Llama API with LiteLLM

  #2734 opened May 1, 2025

• [BUG] Incorrect `supports_response_schema` for OpenRouter models prevents structured output usage

  #2729 opened Apr 30, 2025

• [BUG] Trained data not being used

  #2724 opened Apr 29, 2025

• [FEATURE] @CrewBase can intake agents_config & tasks_config as List

  #2721 opened Apr 29, 2025

• [FEATURE] Decompose complex task into sub-tasks

  #2717 opened Apr 29, 2025

• [BUG] Error when integrating with BrowserUse tool latest version 0.1.40

  #2702 opened Apr 28, 2025

18 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [BUG]

  #2517 commented on Apr 27, 2025 • 0 new comments

• [BUG] Your system has an unsupported version of sqlite3. Chroma

  #2682 commented on Apr 28, 2025 • 0 new comments

• [QUESTION] How to design asynchronous human-in-the-loop Crews running on the backend?

  #2051 commented on Apr 28, 2025 • 0 new comments

• [BUG] Deepseek-v3-250324 Returns Empty Responses When Last Message is from Assistant

  #2657 commented on Apr 28, 2025 • 0 new comments

• [BUG] multimodal not working with gemini 2.5 pro experimental model

  #2642 commented on Apr 28, 2025 • 0 new comments

• [BUG] CrewAI ChromaDB Embedding Dimension Mismatch Issue

  #2464 commented on Apr 29, 2025 • 0 new comments

• [BUG] LLM errors are not getting raised in try exception

  #2487 commented on May 1, 2025 • 0 new comments

• [BUG]

  #2471 commented on May 1, 2025 • 0 new comments

• [FEATURE] How can we simulate a debate or a judge trial with two or multi agents?

  #2236 commented on May 1, 2025 • 0 new comments

• [BUG] Cannot create 'Knowledge': Failed to create or get collection

  #2055 commented on May 1, 2025 • 0 new comments

• [BUG] o4-mini usage raising stop parameter not support exception

  #2661 commented on May 2, 2025 • 0 new comments

• Crewai reset-memories -kn command is not working at all in CLI

  #2678 commented on May 2, 2025 • 0 new comments

• [BUG]

  #2628 commented on May 2, 2025 • 0 new comments

• [BUG]

  #2688 commented on May 2, 2025 • 0 new comments

• [BUG] Docs website search allows empty input

  #2506 commented on May 3, 2025 • 0 new comments

• Introduce more fine control over delegation

  #2362 commented on Apr 28, 2025 • 0 new comments

• Add support for Gemini 2.5 Pro Experimental model

  #2512 commented on May 3, 2025 • 0 new comments

• Fix #2688: Add operation_name parameter to AWS Bedrock embedder configuration

  #2689 commented on Apr 30, 2025 • 0 new comments