[Release] crewaiInc/crewAI 1.14.2: 1.14.2

crewAI 1.14.2 introduces checkpoint management features including resume, diff, and prune commands with lineage tracking and forking capabilities. The release enhances LLM token tracking with reasoning and cache tokens, improves CLI ergonomics for deploy validation and template management, and fixes multiple security vulnerabilities across dependencies. Key improvements include scoped streaming handlers to prevent data contamination, recursive glob for checkpoint discovery, and better HITL (Human-in-the-Loop) resume event handling.

Key Points

• •Checkpoint resume, diff, and prune commands enable better workflow management and discoverability
• •Add `from_checkpoint` parameter to `Agent.kickoff` for seamless checkpoint restoration
• •Checkpoint forking with lineage tracking allows branching and experimentation from saved states
• •Enhanced LLM token tracking now includes reasoning tokens and cache creation tokens for better cost analysis
• •Deploy validation CLI and improved LLM initialization ergonomics streamline deployment workflows
• •Template management commands simplify project template handling and configuration
• •Scoped streaming handlers prevent cross-run chunk contamination and data leakage
• •Recursive glob for JSON checkpoint discovery improves checkpoint file location reliability
• •Multiple security patches address vulnerabilities in authlib, langchain-text-splitters, pypdf, requests, and cryptography
• •Cyclic JSON schema handling in MCP tool resolution prevents infinite loops and crashes

Workflow Diagram

Concepts