1.14.4a1

CrewAI Releases / 4/30/2026

📰 NewsDeveloper Stack & InfrastructureTools & Practical UsageModels & Research

Key Points

  • This release (1.14.4a1) includes multiple bug fixes to improve reliability, including handling crew chat description helpers when LLMs fail and resetting messages/iterations between executor invocations.
  • It fixes trained-agent behavior by forwarding custom trained-agents files through replay/testing in the CLI and honoring them during inference in the agent.
  • The update improves multimodal support by binding task-only agents to crews so that multimodal input files reach the LLM.
  • Guardrail and routing internals were adjusted for safer checkpointing and to prevent self-referential router behavior, including serializing guardrail callables as null and renaming force_final_answer.
  • It also bumps litellm for an SSTI fix and updates documentation with new sandbox tools and guides (E2B, Daytona, Vertex AI workload identity, and You.com MCP).

What's Changed

Bug Fixes

  • Fix crew chat description helpers against LLM failures.
  • Reset messages and iterations between invocations in executor.
  • Forward trained-agents file through replay and test in CLI.
  • Honor custom trained-agents file at inference in agent.
  • Bind task-only agents to crew to ensure multimodal input_files reach the LLM.
  • Serialize guardrail callables as null for JSON checkpointing.
  • Rename force_final_answer in agent_executor to avoid self-referential router.
  • Bump litellm for SSTI fix and ignore unfixable pip CVE.

Documentation

  • Add E2B Sandbox Tools page.
  • Add Daytona sandbox tools documentation.
  • Add Vertex AI workload identity setup guide.
  • Add You.com MCP tools for search, research, and content extraction.
  • Update changelog and version for v1.14.3.

Contributors

@EdwardIrby, @dependabot[bot], @factory-droid-oss, @factory-droid[bot], @greysonlalonde, @lorenzejay, @manisrinivasan2k1, @mattatcha