- Replaced local image paths with CDN links for releases 1.3.11, 1.3.12, 1.4.0, 1.4.1, 1.5.0, and 1.5.2 in the changelog files.
- Updated the changelog entries to ensure consistent image sourcing and improve loading performance.
- Introduced a new Contributors component in the changelog files for releases 1.3.11, 1.3.12, 1.4.1, and 1.5.0 to acknowledge contributors in a consistent format.
- Updated the changelog entries to include a list of contributors for each release, enhancing visibility and appreciation for their contributions.
* chore(docs): remove Blog section from documentation
- Eliminated the Blog tab and its associated group from the documentation structure, streamlining the content for better clarity and focus on relevant topics.
* chore(docs): update image links in changelog and remove obsolete images
- Replaced local image paths with CDN links for releases 1.3.11, 1.3.12, 1.4.0, 1.4.1, and 1.5.0 in the changelog files.
- Deleted outdated image files for releases 1.3.11 and 1.3.12 to streamline documentation and reduce clutter.
* fix(docs): update documentation link for logo reference
- Changed the logo href from the main site to the documentation page for better navigation and user experience.
* chore: update imports
* refactor(mcpagent): update agent creation logic and imports
- Replaced `create_tool_calling_agent` with `create_agent` for improved agent instantiation.
- Simplified the `_create_agent` method by removing the `AgentExecutor` and directly returning the created agent.
- Adjusted type annotations for `_agent_executor` to initialize as `None` instead of using a union type.
* feat(mcpagent): migrate to LangChain 1.0.0 and simplify agent methods
- Updated MCPAgent to utilize LangChain 1.0.0's `create_agent()` for improved agent instantiation.
- Introduced new methods `astream_simplified()` and `run_v2()` that leverage the built-in `astream()` for internal agent loop handling.
- Retained legacy methods `stream()` and `run()` for backward compatibility with manual execution.
- Enhanced documentation to reflect changes and provide clearer usage examples.
* feat(mcpagent): implement dynamic tool updates during agent execution
- Enhanced the MCPAgent to support dynamic tool updates when using server_manager mode.
- Implemented logic to check for tool changes before and during execution, allowing for immediate updates and safe restarts.
- Introduced a maximum restart limit to prevent infinite loops during tool updates.
- Updated documentation to reflect the new behavior and usage of dynamic tool management.
* refactor(mcpagent): clean up imports and streamline logging
- Removed unused imports and organized existing ones for better clarity in `mcpagent.py`.
- Simplified logging statements for consistency and improved readability.
- Enhanced the overall structure of the code by eliminating unnecessary whitespace.
- Updated unit tests to reflect changes in imports and ensure compatibility with the refactored code.
* feat(mcpagent): enhance agent execution with step tracking and middleware
- Introduced `ModelCallLimitMiddleware` to enforce a maximum number of model calls during agent execution.
- Updated the `_consume_and_return` method to return a tuple of the final result and the number of steps taken.
- Enhanced telemetry tracking to include steps taken and execution time for better performance insights.
- Simplified agent creation logic by integrating middleware directly into the `create_agent` function.
- Improved logging for agent execution details, including steps and tool usage.
* fix(mcpagent): add null check for node output messages extraction
- Updated the MCPAgent to include a null check for `node_output` before attempting to access the "messages" key, preventing potential errors during message extraction.
* minor
* feat(mcpagent): filter out ToolMessage from history
- Updated the MCPAgent to exclude ToolMessage instances from being added to the message history, ensuring that only relevant messages are recorded.
- This change helps maintain a cleaner history by avoiding intermediate tool execution details.
* fix(mcpagent): set max_steps assignment
* ruff
* docs
* e2e tests for agent
* key in tests
* fix manager test and split ci tests
* telemetry off for tests
* fix server manager test
* feat(mcpagent): delegate query streaming to remote agent in remote mode
- recreated functionality to the MCPAgent to yield results from a remote agent when operating in remote mode, enhancing the agent's capability to handle queries efficiently.
* fix: agent unit tests
* Print inputs
* rollback, wrong repo
* fix streaming output
* stricter streaming test
* pass observability callbacks
* release notes
* update old release pages
* pyproject version bump
* if you do not want to migrate from langchain classic
* fix tests mock
* Canary Release (#391)
* feat(mcp-server): enhance widget metadata handling and resource domain configuration (#388)
* feat(mcp-server): enhance widget metadata handling and resource domain configuration
- Updated widget title handling to prioritize metadata title over the default name.
- Improved resource domain configuration to support dynamic server base URLs and include wildcard domains for better flexibility.
- Enhanced apps SDK metadata integration by merging additional metadata from widget configurations.
- Refactored HTML template handling for improved clarity and maintainability.
* changeset
* chore: enter canary prerelease mode
* chore(typescript): version packages (canary)
* feat: Enhanced widget integration and tunnel support (#389)
* refactor: streamline widget HTML generation and enhance logging
- Simplified base URL handling in the CLI by removing conditional MCP_URL usage.
- Updated HTML content generation in inspector utilities to improve base tag injection logic.
- Enhanced logging for better debugging and insights into widget content generation.
* feat: add tunnel support to CLI for server exposure
- Introduced a new `--tunnel` option in the CLI to expose the server through a tunnel.
- Implemented `startTunnel` helper function to manage tunnel creation and URL retrieval.
- Enhanced server startup logic to utilize the tunnel URL when available, improving accessibility for remote connections.
- Updated logging to provide feedback on tunnel status and URLs during server initialization.
* feat(inspector): enhance tunnel URL handling and update dependencies
- Added support for preserving the tunnel URL in navigation across inspector components.
- Introduced a new `tunnelUrl` state in the Inspector context to manage tunnel URL across the application.
- Updated various components to utilize the tunnel URL, ensuring seamless integration with the new functionality.
- Added `@radix-ui/react-popover` as a dependency in package.json for improved UI interactions.
- Updated pnpm-lock.yaml to reflect changes in dependency versions, including zod.
* feat(mcp): enhance widget integration and server handling
- Updated the CLI to fetch the MCP server from the new `/mcp` endpoint instead of `/inspector`.
- Improved tunnel establishment logging to include the new `/mcp` path.
- Enhanced HTML generation in the MCP server to replace relative paths for widget scripts and styles with absolute URLs.
- Added a script to the HTML head for dynamic file retrieval based on the server base URL, improving widget loading efficiency.
* disable tunnel in dev for now (since vite does not build assets
* fix(inspector): improve header handling in registerInspectorRoutes
- Updated header processing to exclude 'accept-encoding' to prevent compression issues.
- Added explicit request for uncompressed responses by setting 'Accept-Encoding' to 'identity'.
- Enhanced response header forwarding by omitting compression-related headers to ensure accurate body state representation.
* chore(typescript): version packages (canary)
* Feat/create mcp use app/templates (#390)
* readmes
* readme addition changeset
* chore(typescript): version packages (canary)
* format docs
---------
Co-authored-by: Enrico Toniato <2827496+tonxxd@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* release(typescript): @mcp-use/inspector 0.4.7 + mcp-use 1.1.8 + create-mcp-use-app 0.4.8 + @mcp-use/cli 2.1.20
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* Merge commit from fork
* fix: format
* feat(mcpagent): migrate to LangChain 1.0.0 and simplify agent methods
- Updated MCPAgent to utilize LangChain 1.0.0's `create_agent()` for improved agent instantiation.
- Introduced new methods `astream_simplified()` and `run_v2()` that leverage the built-in `astream()` for internal agent loop handling.
- Retained legacy methods `stream()` and `run()` for backward compatibility with manual execution.
- Enhanced documentation to reflect changes and provide clearer usage examples.
* feat(mcpagent): implement dynamic tool updates during agent execution
- Enhanced the MCPAgent to support dynamic tool updates when using server_manager mode.
- Implemented logic to check for tool changes before and during execution, allowing for immediate updates and safe restarts.
- Introduced a maximum restart limit to prevent infinite loops during tool updates.
- Updated documentation to reflect the new behavior and usage of dynamic tool management.
* feat(mcpagent): enhance agent execution with step tracking and middleware
- Introduced `ModelCallLimitMiddleware` to enforce a maximum number of model calls during agent execution.
- Updated the `_consume_and_return` method to return a tuple of the final result and the number of steps taken.
- Enhanced telemetry tracking to include steps taken and execution time for better performance insights.
- Simplified agent creation logic by integrating middleware directly into the `create_agent` function.
- Improved logging for agent execution details, including steps and tool usage.
* minor
* ruff
* format
---------
Co-authored-by: Luigi Pederzani <luigipederzani@gmail.com>
Co-authored-by: Enrico Toniato <2827496+tonxxd@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Vincenzo Reina <vincenzoreinaa05@gmail.com>
Luigi Pederzani