Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
http://www.langflow.org
a276c2de48* refactor: Simplify flow execution and update input handling in MCP API - Replaced `InputValueRequest` with `SimplifiedAPIRequest` for cleaner input management. - Updated flow execution logic to utilize `simple_run_flow`, enhancing clarity and performance. - Removed unnecessary background task handling and streamlined progress updates. - Improved message collection from flow outputs, ensuring robust handling of results. * fix: Add error handling for tool execution in MCP API - Implemented a try-except block around the flow execution to catch and handle exceptions gracefully. - Enhanced message collection logic to ensure that errors during tool execution are communicated back as text content. - Improved robustness of the `handle_call_tool` function by ensuring all potential errors are captured and reported. * fix: Improve error messaging in tool execution for MCP API - Updated error handling in the `handle_call_tool` function to provide more descriptive error messages. - Changed the error message format to include the flow name, enhancing clarity for debugging purposes. - Ensured that all exceptions during tool execution are captured and reported as text content. * refactor: Enhance message and result handling in handle_call_tool - Improved the logic for processing outputs in the `handle_call_tool` function to handle messages and results more comprehensively. - Streamlined the collection of text content from both messages and results, ensuring all relevant outputs are captured. - Enhanced robustness by ensuring that all outputs are processed, regardless of their structure. * refactor: Improve progress notification handling in handle_call_tool - Updated the logic for progress task creation in the `handle_call_tool` function to ensure it only initializes when progress notifications are enabled and a progress token is present. - Enhanced the cancellation and exception handling of the progress task to prevent potential errors when it is not created. - Improved overall robustness of the function by ensuring that progress updates are managed correctly based on the current context. * refactor: Streamline flow execution and message handling in ProjectMCPServer - Replaced `InputValueRequest` with `SimplifiedAPIRequest` for improved input management. - Updated flow execution to utilize `simple_run_flow`, enhancing clarity and performance. - Refined progress notification handling to ensure tasks are only created when necessary. - Improved message collection from flow outputs, ensuring robust handling of both messages and results. - Enhanced error handling during tool execution to provide clearer feedback on failures. * refactor: enhance progress update handling in ProjectMCPServer Updated the send_progress_updates function to accept a progress token as an argument, improving its flexibility. Adjusted the task cancellation logic to use asyncio.gather for better exception handling. This change aims to streamline progress notifications when enabled. * refactor: add group_outputs property to message configurations in starter projects |
||
|---|---|---|
| .cursor/rules | ||
| .devcontainer | ||
| .github | ||
| .vscode | ||
| deploy | ||
| docker | ||
| docker_example | ||
| docs | ||
| scripts | ||
| src | ||
| test-results | ||
| .composio.lock | ||
| .env.example | ||
| .eslintrc.json | ||
| .gitattributes | ||
| .gitignore | ||
| .pre-commit-config.yaml | ||
| CODE_OF_CONDUCT.md | ||
| CONTRIBUTING.md | ||
| DEVELOPMENT.md | ||
| eslint.config.js | ||
| LICENSE | ||
| Makefile | ||
| pyproject.toml | ||
| README.md | ||
| render.yaml | ||
| uv.lock | ||
Langflow is a powerful tool for building and deploying AI-powered agents and workflows. It provides developers with both a visual authoring experience and a built-in API server that turns every agent into an API endpoint that can be integrated into applications built on any framework or stack. Langflow comes with batteries included and supports all major LLMs, vector databases and a growing library of AI tools.
✨ Highlight features
- Visual Builder to get started quickly and iterate.
- Access to Code so developers can tweak any component using Python.
- Playground to immediately test and iterate on their flows with step-by-step control.
- Multi-agent orchestration and conversation management and retrieval.
- Deploy as an API or export as JSON for Python apps.
- Observability with LangSmith, LangFuse and other integrations.
- Enterprise-ready security and scalability.
⚡️ Quickstart
Langflow works with Python 3.10 to 3.13.
Install with uv (recommended)
uv pip install langflow
Install with pip
pip install langflow
📦 Deployment
Self-managed
Langflow is completely open source and you can deploy it to all major deployment clouds. Follow this guide to learn how to use Docker to deploy Langflow.
Fully-managed by DataStax
DataStax Langflow is a full-managed environment with zero setup. Developers can sign up for a free account to get started.
⭐ Stay up-to-date
Star Langflow on GitHub to be instantly notified of new releases.
👋 Contribute
We welcome contributions from developers of all levels. If you'd like to contribute, please check our contributing guidelines and help make Langflow more accessible.