Tool Use

LangChain v1.3.0a1 is an alpha release featuring stream_events(version='v3') protocol integration into create_agent, HITL middleware enhancements with a 'respond' decision, and performance optimizations including reduced agent state inlining and LRU caching for type resolution. The release includes updates across multiple partner packages (Fireworks, Anthropic, OpenAI), dependency bumps for security (Pygments CVE-2026-4539, cryptography), and improvements to token counting, middleware initialization speed, and agent configuration. Key infrastructure changes include content-block-centric streaming (v2), dynamic tool registration via middleware, and enhanced LangSmith integration metadata.

crewAI v1.14.5a1 introduces the `restore_from_state_id` kickoff parameter for state management and enhances the ExaSearchTool with highlights while renaming it from EXASearchTool. Bug fixes address missing crewai pin sites in the release flow and ensure proper skills loading events for traces. This alpha release includes documentation updates and contributions from five team members.

langchain-core version 1.4.0a1 is an alpha release featuring the new stream_events(version='v3') protocol for improved event streaming. The release includes fixes for structured inputs preservation in tool runs, optional removal parameters in deprecation warnings, and batch size validation to prevent infinite loops. Key improvements focus on streaming capabilities, tracer metadata handling, and various bug fixes across core functionality.

langchain-core version 1.4.0a2 is released with significant improvements to event streaming, tracing, and core functionality. Key updates include unwinding the _AsyncEventsResult hybrid implementation, introducing stream_events(version='v3') protocol, and preserving structured inputs on tool runs in tracers. The release includes numerous bug fixes, performance optimizations, and dependency updates across the LangChain ecosystem.

Google ADK Python v1.32.0 introduces significant enhancements to agent development capabilities, including native OpenTelemetry metrics, Anthropic thinking blocks support, improved authentication mechanisms (2LO/3LO via GcpAuthProvider), and express mode deployment. The release addresses critical security vulnerabilities (RCE via YAML, SSRF), fixes parallel tool execution issues, and improves analytics with BigQuery LLM cache metadata logging and credential isolation to prevent race conditions.

OpenAI Agents Python v0.15.0 introduces explicit model refusal handling through a new `ModelRefusalError` exception, replacing previous behavior where refusals were treated as empty output or caused retry loops. The release includes a `model_refusal` error handler that allows developers to gracefully handle refusals without raising exceptions. For structured-output agents, handlers can return values matching the output schema for validation. Additional updates include documentation improvements and dependency bumps.

Claude Code v2.1.126 introduces gateway model listing, project purge functionality, improved OAuth authentication for WSL/SSH/containers, and numerous bug fixes. Key improvements include permission bypass options, better shell detection on Windows, security fixes for managed settings, and stability enhancements for streaming, image handling, and remote sessions. The release also refines telemetry events, fixes UI issues, and resolves edge cases in tool availability and clipboard handling.

langchain-fireworks v1.2.1 is a patch release that fixes multimodal content block translation for chat completions, updates dependencies (aiohttp and requests), refreshes model profile data, and includes core streaming improvements. The release addresses compatibility issues with minimum core versions and updates documentation references.

LangChain version 1.2.17 has been released with a new feature adding a `respond` decision to the HITL (Human-in-the-Loop) middleware. This update enhances the middleware's decision-making capabilities by allowing systems to respond directly within the human-in-the-loop workflow. The release builds on version 1.2.16 with focused improvements to middleware functionality.

OpenClaw v2026.4.29-beta.4 introduces significant enhancements across messaging, memory, provider coverage, and reliability. Key additions include active-run steering by default, a people-aware wiki with provenance views, NVIDIA provider onboarding, and improved gateway/channel reliability. Security improvements include OpenGrep scanning and sharper GHSA triage policies, while multiple channel fixes address Slack, Telegram, Discord, WhatsApp, and Teams edge cases.

OpenClaw v2026.4.29 is a major release introducing active-run steering for messaging, people-aware memory wiki with provenance tracking, expanded provider coverage (NVIDIA, Bedrock), and enhanced reliability across gateway, channels, and security. Key improvements include opt-in follow-up commitments with heartbeat delivery, per-conversation Active Memory filters, faster model manifest paths, and comprehensive channel fixes for Slack, Telegram, Discord, WhatsApp, and Teams. Security enhancements add OpenGrep scanning and safer exec/pairing handling.

crewAI v1.14.4 introduces enhanced persistence customization, expanded cloud provider support (Azure OpenAI, Vertex AI), and new research tools (Tavily, You.com MCP). The release includes 12 bug fixes addressing JSON handling, message preservation, credential forwarding, and multimodal input support. Documentation updates cover E2B and Daytona sandbox tools.

OpenClaw v2026.4.29-beta.2 introduces significant enhancements across messaging, memory, provider coverage, and reliability. Key additions include active-run steering for messaging, a people-aware wiki with provenance tracking, NVIDIA provider onboarding, and improved channel resilience across Slack, Telegram, Discord, and others. Security improvements include OpenGrep scanning and stricter tool-scope handling, while operational reliability focuses on slow-host startup diagnostics and session recovery.

langchain-perplexity v1.2.0 introduces PerplexityEmbeddings support and overhauls the integration with the official Perplexity SDK and Search API. The release includes content-block-centric streaming improvements, dependency updates (langsmith, pillow, pytest), security patches (CVE-2026-4539 via pygments), and standardized integration testing across partners. Multiple maintenance updates ensure compatibility with langchain-core 1.2.21+ and improve code quality through type checking and documentation enhancements.

LangChain v1.2.16 release includes performance optimizations for agent state handling and streaming, new content-block-centric streaming in core, dependency updates, and documentation corrections. Key improvements focus on reducing overhead in tool dispatch, adding LRU caching for type resolution, and enhancing streaming metadata efficiency. The release also bumps related packages including Anthropic 1.4.2, OpenAI 1.2.1, and LangChain Core 1.3.0.

OpenClaw v2026.4.27 introduces major enhancements across Codex Computer Use setup, new provider integrations (DeepInfra, Tencent Yuanbao, QQBot), and a shift toward manifest-first plugin metadata architecture. Key improvements include GPU passthrough for Docker sandboxes, authenticated node presence tracking for iOS/Android, outbound proxy routing, and comprehensive reliability fixes across Telegram, Slack, and gateway startup. The release emphasizes reducing boot overhead through manifest-driven plugin discovery and model catalogs.

crewAI v1.14.4a1 is a pre-release update focusing on bug fixes and documentation improvements. Key fixes address LLM chat failures, message state management in executors, trained-agent file handling, multimodal input support, and JSON serialization issues. The release includes new documentation for sandbox tools (E2B, Daytona), Vertex AI setup, and You.com MCP integration, plus security updates to litellm.

langchain-tests version 1.1.7 is released with several improvements including a hotfix for minimum core version requirements, new content-block-centric streaming capabilities (v2), and dependency updates. Key updates include bumping langsmith from 0.6.3 to 0.7.31 and pytest to 9.0.3. This release focuses on stability, testing infrastructure improvements, and enhanced streaming functionality.

OpenAI Agents Python v0.14.8 is a maintenance release that fixes two critical issues: preserving MCP re-export import errors and properly delimiting sandbox prompt instruction sections. The release includes documentation improvements and welcomes a new contributor to the project.

Claude Code v2.1.123 is a patch release that fixes a critical OAuth authentication bug. When the experimental betas feature flag was disabled via the `CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1` environment variable, OAuth authentication would enter a 401 retry loop, preventing successful login. This fix resolves the authentication failure and allows users with experimental features disabled to authenticate properly.