Skills & Tools

langchain-perplexity version 1.4.1 is released with several bug fixes and dependency updates. Key improvements include fixing the Responses API input format by adding type="message" attributes, preserving caller extra_body parameters, and using the correct supported Responses API model. The release also includes dependency bumps for pillow, vcrpy, and langsmith, along with documentation updates and package version tracking enhancements.

Claude Code v2.1.236 introduces the `ANTHROPIC_DEFAULT_MODEL` environment variable for persistent model selection, adds cross-session idle notifications via `notify_when_idle`, and enhances sandbox security with precedence-based wildcard read-deny rules on macOS. The release fixes 20+ bugs affecting clipboard operations, fullscreen rendering, subprocess handling, and terminal UI consistency, while improving auto mode classification, startup performance, and accessibility with screen reader support.

langchain-core version 1.6.0 introduces bug fixes, performance improvements, and new features including standard model exception types and enhanced tool schema handling. Key improvements address serialization issues, Windows portability, forward reference resolution, and lazy loading of transformers. The release focuses on stability and developer experience with better error handling and documentation clarification.

langchain-core version 1.5.4 is a maintenance release focused on bug fixes and compatibility improvements. Key updates include Pydantic 2.14 compatibility, fixes for StructuredPrompt mutation issues, proper handling of tool arguments and schemas, and improvements to streaming and callback mechanisms. The release addresses 13 specific issues across core functionality, infrastructure, and documentation.

LangChain v1.3.15 release includes 30+ improvements focused on middleware enhancements, bug fixes, and dependency updates. Key additions include trace_policy exposure on AgentMiddleware, state_schema parameter for wrap_tool_call, and LangSmith provider integration in init_chat_model. The release addresses critical issues with history preservation, structured output handling, HITL approval gates, and tool call management while maintaining security through dependency bumps.

Claude Code v2.1.228 is a maintenance release fixing critical bugs in interactive sessions, Git integration on Windows, model switching, cross-session messaging, and self-hosted runners. Key improvements include hardened skill syncing from claude.ai, better Vertex AI credential handling, improved UI feedback during compaction, and updated file-writing permissions for newer models. The release addresses 17 distinct issues spanning session stability, remote control functionality, plugin caching, and settings management.

Google ADK Python v1.38.0 release introduces safety settings forwarding to the Live API and addresses critical security and stability issues. Key improvements include preventing shell execution vulnerabilities in ReadFileTool, excluding problematic nltk dependency versions, fixing Live API agent transfers, and enhancing evaluation error handling. This release focuses on security hardening and reliability improvements for agent development.

crewAI 1.15.12 introduces URLReadTool for arbitrary URL reading, unifies resource scaffolding under a single `crewai create` command, and adds app metadata to platform action tools. Bug fixes clarify conversational route/handler name collision errors, with updated documentation for the new unified CLI. No breaking changes are included in this release.

crewAI v1.15.11 introduces enhanced telemetry tracking for interception-hook dispatches and project_id linking for enterprise accounts, adds IBM Db2 search tool support, and surfaces AMP (Agent Management Platform) in documentation. The release includes critical security updates for aiohttp and cryptography dependencies, fixes CodeQL alerts, and improves CI/CD pipeline efficiency by skipping code checks for documentation-only PRs.

Google ADK Python v2.6.1 release introduces enhanced telemetry capabilities for CLI operations, including parent terminal grouping, TTL pruning, and TTY connectivity tracking. Bug fixes address early telemetry recording for long-running web servers and eliminate redundant tool confirmation validation. This update improves observability and performance of the ADK CLI toolchain.

crewAI v1.15.10 introduces skill usage event collection capabilities, enabling better tracking and monitoring of agent skill utilization. The release includes documentation updates removing migrated AMP content and refreshing security reporting guidelines. This minor version update maintains stability while adding observability features for skill performance analysis.

Google ADK Python v2.6.0 introduces significant enhancements for agent development, including per-invocation auth headers for A2A, new evaluation samples, OAuth2 improvements, and telemetry capabilities. Key additions include the ReflectAndRetryModelPlugin for self-healing, OCI Generative AI provider integration, Eventarc Advanced toolset, and ManagedAgent instruction field support. The release also improves CLI functionality with telemetry consent checks, command logging, and enhanced error handling across multiple agent frameworks.

crewAI 1.15.9 release introduces improved error handling by surfacing tool failures explicitly rather than masking them as successes, adds FlowFailedEvent emission for better flow execution monitoring, and implements progressive disclosure for skills. The release includes documentation updates for the previous version and contributions from multiple team members.

OpenAI Agents Python SDK v0.19.0 introduces Programmatic Tool Calling, allowing supported models to generate JavaScript for coordinating tools with per-tool caller restrictions and structured outputs. The release adds a public decorators module with a shorter @tool alias, standardizes SDK configuration to accept both typed objects and dictionaries, hardens sensitive logging across all components, and improves compatibility with AnyLLM, LiteLLM, and Chat Completions while preserving session history on retries.

CrewAI 1.15.7a1 is a patch release addressing critical bug fixes and observability improvements. Key updates include fixing registry skills resolution through CrewAI+ client, recovering from GPT-5.6 tool errors, enabling tool calling on the Responses API, and preventing 404 errors for responses-only models. The release also adds skill usage event emission for better runtime observability and includes a security patch for a bedrock-agentcore dependency vulnerability.

CrewAI v1.15.7 release includes critical bug fixes for registry skills resolution, GPT-5.6 tool compatibility, and responses-only model routing. The update enhances observability by emitting skill usage events at runtime and addresses a security vulnerability in bedrock-agentcore. This patch ensures improved tool calling functionality across different API paths and model types.

crewAI v1.15.6 is a maintenance release focusing on bug fixes and stability improvements. Key fixes include correcting Anthropic preview tool-use block detection, preserving strict tool schema property names, ensuring execution_end hooks dispatch on failed executions, handling async agent loading, and resolving dependency issues. The release includes documentation updates and contributions from six team members.

Google ADK Python v2.5.0 introduces significant enhancements for agent development, including Cloud Run sandbox execution, MCP server integration, strict input validation, and improved workflow state management. Key additions include Voice Activity Detection support, HITL resumption capabilities, binary document parsing, and enhanced error handling callbacks. The release also addresses security concerns with module blocking, mTLS support, and comprehensive bug fixes across event compaction, model caching, and agent orchestration.

crewAI version 1.15.5 introduces authenticated skill registry downloads, enhancing security for accessing external skills. This release includes documentation updates for the previous version's snapshot and changelog. The update is a minor release focused on improving the authentication mechanism for skill registry operations.

Claude Code v2.1.215 changes how verification and code review skills are invoked. Previously, Claude automatically ran `/verify` and `/code-review` skills, but now these must be explicitly invoked by the user. This gives users more control over when these operations execute, reducing unnecessary processing and improving efficiency.