Tool Use

OpenClaw v2026.7.2-beta.5 introduces major reliability and safety improvements including crash-recoverable state management with quarantine stores, durable channel delivery across multiple platforms, session rewind/branching capabilities, interactive MCP Apps with dashboards, and structured question/approval workflows. The release adds support for new models (Claude Opus 5, Kimi K3, GPT Live), native apps for macOS/Linux/Wear OS, meeting integrations (Teams, Zoom, Google Meet), and guided local inference setup. Extensive security fixes address authorization, SQLite data safety, channel delivery persistence, and session lifecycle management.

langchain-fireworks version 1.5.2 has been released with updates focused on model profile data refreshes. The release includes a primary change to the fireworks module along with three separate model profile data refresh commits. This is a maintenance release that ensures the Fireworks integration stays current with the latest model configurations and profiles.

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.

langchain-anthropic version 1.5.2 has been released with support for Claude Opus 5. This update builds on the previous 1.5.1 release and includes new model capabilities for the Anthropic integration within the LangChain framework. The release enables developers to leverage the latest Claude Opus 5 model through LangChain's unified interface.

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.

langchain-anthropic version 1.5.1 has been released with three key improvements. The update adds support for LangSmith gateway configuration through environment variables, enabling structured output for Claude Opus 4.8 model, and includes general release updates. This version enhances integration capabilities and model compatibility for Anthropic-based LangChain applications.

langchain-core version 1.5.1 is released with three key improvements: support for LangSmith gateway configuration via environment variables across Anthropic, Fireworks, and OpenAI integrations, and a fix for BaseTool token counting that leverages tool_call_schema caching in the count_tokens_approximately method. This patch release enhances gateway flexibility and optimizes token counting performance.

langchain-fireworks version 1.5.1 has been released with a key feature addition enabling LangSmith gateway support through environment variables. This update improves integration capabilities across Anthropic, Fireworks, and OpenAI providers by allowing users to configure gateway access via environment configuration rather than code changes. The release builds on version 1.5.0 with enhanced flexibility for enterprise deployments and monitoring workflows.

langchain-openai version 1.4.1 has been released with three key updates: support for LangSmith gateway configuration through environment variables across Anthropic, Fireworks, and OpenAI integrations, and a correction to the gpt-5.3-chat-latest profile. This release enables better integration with LangSmith monitoring and fixes a profile naming issue in the OpenAI provider.

Claude Code v2.1.218 introduces significant improvements to code review workflows, accessibility features, and bug fixes. Key changes include running `/code-review` as a background subagent to reduce conversation clutter, enhanced screen-reader support with deletion announcements, and fixes for critical issues like Windows path corruption and multi-line paste handling. The release also improves error handling, trust dialogs, and agent configuration validation while refining auto-mode behavior and MCP server connectivity reporting.

langchain-openrouter version 0.2.7 has been released with updates since 0.2.6. The release primarily consists of routine maintenance work including multiple model profile data refreshes and dependency lockfile updates. No breaking changes or major feature additions are documented in this release.

Claude Code v2.1.217 introduces emoji shortcode autocomplete, improves transcript reliability with failure warnings, and fixes critical issues including memory leaks, Windows auto-update failures, and security vulnerabilities in session isolation. The release also enhances subagent management with concurrency caps and nesting controls, while addressing telemetry routing, screen reader accessibility, and various edge cases in session handling and transcript management.

langchain-anthropic version 1.5.0 introduces support for `reasoning_effort` as a standard chat model parameter, enabling more granular control over Claude's reasoning capabilities. The release includes fixes for tool recognition with the `advisor_` prefix, improved documentation following Sonnet 5 release, and various dependency updates. Key improvements focus on enhancing the Anthropic integration's compatibility with extended thinking features and maintaining consistency across the LangChain ecosystem.

langchain-fireworks version 1.5.0 introduces a new `reasoning_effort` parameter as a standard chat model feature, enabling more granular control over model reasoning behavior. The release includes dependency updates to langsmith (0.10.6), security patches for Dependabot vulnerabilities, and maintenance improvements across the Fireworks integration library.

langchain-openai version 1.4.0 has been released with several updates including a Pillow dependency bump, addition of `reasoning_effort` as a standard chat model parameter for enhanced model control, and refreshed model profile data. This release builds on version 1.3.5 with improvements to core functionality and dependency management.

langchain-xai version 1.3.0 introduces the `reasoning_effort` parameter as a standard chat model feature, along with numerous dependency updates and security patches. The release includes fixes for unsupported parameters, improved model profile data, and enhanced streaming capabilities with content-block-centric v2 support. Key updates involve langsmith, pytest, aiohttp, and other core dependencies, with emphasis on security vulnerabilities and compatibility improvements.

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.

langchain-core version 1.5.0 has been released with new features and dependency updates. The major addition is support for `reasoning_effort` as a standard chat model parameter, enabling more granular control over model reasoning behavior. Dependencies have been updated including soupsieve (2.8 → 2.8.4) and mistune (3.2.1 → 3.3.0) for improved security and functionality.