Tool Use

Claude Code v2.1.243 introduces significant improvements to usage monitoring, model management, and authentication. Key additions include a Loops breakdown in `/usage` for tracking runaway tasks, a `modelPicker` setting for curating model selections, prompt cache TTL controls, organization-managed pricing, and keyless sign-in via Anthropic Console. The release also includes numerous bug fixes for MCP servers, auto mode, session management, and performance optimizations reducing binary size by ~75% and memory usage by 40-70 MB.

OpenClaw v2026.8.1-beta.3 introduces GPT-5.6 reasoning support across multiple model variants (Sol, Terra, Luna, Ultra) and enhances the Control UI with streamlined first-run setup that integrates model verification into Custodian and optional channel configuration. Key infrastructure improvements include Puppeteer-compatible CDP relay support for Chrome sessions, explicit Gateway lifecycle supervision with verified restart handoff, and compact SQLite backup/restore commands. All 89 official npm plugins have been verified and published with complete integrity metadata.

Claude Code v2.1.239 is a major release featuring improved cost estimation with US-only inference premiums, enhanced cloud session plugin management, Python 3.1.x migration tools, and extensive bug fixes across Bedrock, Alpine/musl builds, proxy handling, IDE integration, and UI rendering. The update addresses 50+ issues including streaming fixes, race conditions, MCP server reliability, and keyboard/mouse input handling in fullscreen mode.

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 v2.1.238 introduces keybinding customization, plugin marketplace enhancements with header helpers, improved self-hosted runner capabilities, and numerous bug fixes. Key improvements include memory leak fixes for long sessions, better Remote Control resilience, cross-session messaging enhancements, and MCP server reliability improvements. The release focuses on stability, performance, and user experience across CLI, Desktop, and remote deployment scenarios.

langchain-fireworks version 1.6.0 introduces document reranking capabilities and improves tool call handling by filtering invalid v1 content. The release includes standard model exception types, refreshed model profile data, and dependency updates for langsmith and aiohttp. Key fixes address repeated service_tier parameters and enhance overall stability.

langchain-anthropic version 1.6.1 was released with a critical bug fix that filters invalid tool calls from v1 content. This patch addresses issues with tool call handling in the Anthropic integration for LangChain, improving reliability when processing tool invocations. The release maintains backward compatibility while resolving edge cases in content processing.

LangChain v1.3.16 introduces standard model exception types for improved error handling, adds support for custom token counters in ContextEditingMiddleware, and fixes critical issues with exception re-raising in ModelRetryMiddleware and schema ordering preservation. This release enhances the framework's robustness in handling errors and managing context within middleware components.

crewAI v1.15.17 introduces declarative conversational flows with opt-in conversational mode, improves tool handling with AMP slug support and oversized message chunking, and fixes critical bugs including MCP server configuration, agent scope closure, tool error attribution, and OpenAI Responses API compatibility. The release enhances security with SSRF checks per redirect hop and resolves native tool call issues.

Claude Code v2.1.237 introduces two key improvements: a fix for prompt caching compatibility with LLM gateways and custom base URLs, and a new "Concise" output style that prioritizes results over preamble. The Concise mode maintains thorough work while reducing narrative overhead, available in the Output style configuration.

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-anthropic version 1.6.0 introduces standard model exception types for improved error handling across the LangChain framework. The release includes infrastructure improvements for LangSmith gateway support in CI/CD pipelines and fixes a grep search scope issue in the Anthropic integration. This update enhances reliability and consistency in exception handling for Anthropic-based LangChain applications.

langchain-openai version 1.6.0 introduces standard model exception types for improved error handling across the core framework. The release includes a fix that raises clear, informative errors when unexpected response types are encountered in the `_create_chat_result` method. These changes enhance debugging and error transparency for OpenAI integration users.

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.

OpenAI Agents Python v0.22.0 is a minor release focused on runtime hardening and provider configuration improvements. Key changes include redacting blocked tool outputs from replay state, rejecting conflicting provider options when explicit clients are used, isolating usage accounting between RunState checkpoints, and expanding handoff agents in generated graphs. Applications combining explicit `openai_client` with `organization` or `project` parameters must migrate those values to the `AsyncOpenAI` client constructor.

Claude Code v2.1.235 introduces a spellcheck feature for prompt input and delivers numerous bug fixes and UX improvements. Key fixes address prompt caching invalidation, markdown list alignment, UI highlight positioning, permission dialogs, and Vim mode behavior. Performance improvements reduce memory/CPU usage for cloud sessions, and several edge cases in dialogs, slash commands, and cross-session messaging are resolved.

langchain-openai version 1.5.2 is released with several bug fixes and feature enhancements. Key improvements include preserving reasoning item boundaries in OpenAI responses, extracting gateway metadata from response headers, and adding support for o-series models in token counting. The release addresses compatibility issues and enhances the integration between LangChain and OpenAI's API.

langchain-openai version 1.5.2a1 is released with support for extracting gateway metadata from response headers and improved handling of o-series models in token counting. The release includes fixes for streamed encrypted reasoning, invalid tool calls, and context window errors, along with support for the OpenAI 3.0 SDK. Recent versions have added features like explicit prompt caching, ChatGPT OAuth-backed models, and content-block-centric streaming improvements.

Claude Code v2.1.234 introduces environment variable configuration for project directories, GitLab merge request integration, automatic session continuation at usage limits, and significant security hardening against Windows NT-namespace path exploits. The release includes numerous bug fixes for session management, permission handling, markdown rendering, and Remote Control functionality, plus UX improvements for fullscreen mode and mid-turn command execution.

LangChain Core version 1.5.6 has been released with a focus on improving observability and tracing capabilities. The primary feature addition incorporates gateway metadata into traces, enhancing the ability to track and monitor AI agent interactions. This release builds on version 1.5.5 with targeted improvements to the core tracing infrastructure.