Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
http://www.langflow.org
933198d8be* feat: add dependency analysis utilities for custom components - Introduced `dependency_analyzer.py` to analyze and classify dependencies in Python code. - Implemented functions to extract import information and categorize dependencies as standard library, local, or external. - Enhanced `build_component_metadata` to include dependency analysis results in component metadata. - Added unit tests to validate the functionality of the dependency analysis features. * refactor: streamline dependency analysis by filtering out stdlib and local imports - Updated `dependency_analyzer.py` to focus on external dependencies only, removing standard library and local imports from analysis results. - Simplified the `DependencyInfo` class by eliminating unnecessary attributes and adjusting the deduplication logic. - Modified `build_component_metadata` to reflect changes in dependency structure, removing counts for stdlib and local dependencies. - Enhanced unit tests to validate the new filtering behavior and ensure no duplicates in external dependencies. * feat: update starter project metadata with dependency information - Added dependency sections to multiple starter project JSON files, specifying required packages and their versions. - Included `langflow` version `1.5.0.post1` and other relevant dependencies such as `orjson`, `fastapi`, and `pydantic` across various projects. - Enhanced project metadata to improve clarity on external dependencies for better maintainability and user guidance. * ⚡️ Speed up function `_classify_dependency` by 7,582% in PR #9192 (`add-deps-metadata`) (#9193) Co-authored-by: codeflash-ai[bot] <148906541+codeflash-ai[bot]@users.noreply.github.com> * fix: ensure distribution version is returned correctly in `_get_distribution_version` - Updated `_get_distribution_version` function to return the distribution version after successfully retrieving it, addressing a potential issue where `None` could be returned prematurely. * fix: improve distribution version lookup in `_get_distribution_version` * fix: handle distribution version lookup exceptions more gracefully * [autofix.ci] apply automated fixes * [autofix.ci] apply automated fixes * [autofix.ci] apply automated fixes (attempt 2/3) * fix(apply_tweaks): skip tweaks to code field and log warning (#9467) * fix: add security warning for overriding code field in tweaks * test: add tests for preventing code field overrides in tweaks * ref: Refactor vectorstore components structure (#9486) * Refactor vectorstore components structure Moved vectorstore components for Chroma, ClickHouse, Couchbase, DataStax, Elastic, Milvus, MongoDB, Pinecone, Qdrant, Supabase, Upstash, Vectara, and Weaviate into dedicated subfolders with __init__.py files for each. Updated Redis vectorstore implementation to reside in redis.py and removed the old vectorstores/redis.py. Adjusted starter project JSONs and frontend constants to reflect new module paths and sidebar entries for these vectorstores. * Refactor vectorstore components and add lazy imports Moved Datastax-related files from vectorstores to a dedicated datastax directory. Added lazy import logic to __init__.py files for chroma, clickhouse, couchbase, elastic, milvus, mongodb, pinecone, qdrant, supabase, upstash, vectara, and weaviate components. Cleaned up vectorstores/__init__.py to only include local and faiss components, improving modularity and import efficiency. * [autofix.ci] apply automated fixes * Refactor vectorstore components structure Moved FAISS, Cassandra, and pgvector components to dedicated subdirectories with lazy-loading __init__.py files. Updated imports and references throughout the backend and frontend to reflect new locations. Removed obsolete datastax Cassandra component. Added new sidebar bundle entries for FAISS, Cassandra, and pgvector in frontend constants and style utilities. * Add lazy imports and Redis chat memory component Refactored the Redis module to support lazy imports for RedisIndexChatMemory and RedisVectorStoreComponent, improving import efficiency. Added a new redis_chat.py file implementing RedisIndexChatMemory for chat message storage and retrieval using Redis. * Fix vector store astra imports * Revert package lock changes * More test fixes * Update test_vector_store_rag.py * Update test_dynamic_imports.py * Update vector_store_rag.py * Update test_dynamic_imports.py * Refactor the cassandra chat component * Fix frontend tests for bundle * Mark Local DB as legacy * Update inputComponent.spec.ts * [autofix.ci] apply automated fixes --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Eric Hare <ericrhare@gmail.com> Co-authored-by: Carlos Coelho <80289056+carlosrcoelho@users.noreply.github.com> * feat: add dependencies metadata to starter projects * feat: add caching for packages_distributions to improve performance * refactor: update test descriptions and remove unused imports in metadata tests --------- Co-authored-by: codeflash-ai[bot] <148906541+codeflash-ai[bot]@users.noreply.github.com> Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Edwin Jose <edwin.jose@datastax.com> Co-authored-by: Eric Hare <ericrhare@gmail.com> Co-authored-by: Carlos Coelho <80289056+carlosrcoelho@users.noreply.github.com> |
||
|---|---|---|
| .cursor | ||
| .devcontainer | ||
| .github | ||
| .vscode | ||
| deploy | ||
| docker | ||
| docker_example | ||
| docs | ||
| scripts | ||
| src | ||
| test-results | ||
| .coderabbit.yaml | ||
| .composio.lock | ||
| .dockerignore | ||
| .env.example | ||
| .eslintrc.json | ||
| .gitattributes | ||
| .gitignore | ||
| .pre-commit-config.yaml | ||
| CODE_OF_CONDUCT.md | ||
| codecov.yml | ||
| CONTRIBUTING.md | ||
| DEVELOPMENT.md | ||
| LICENSE | ||
| Makefile | ||
| Makefile.frontend | ||
| pyproject.toml | ||
| README.md | ||
| RELEASE.md | ||
| render.yaml | ||
| SECURITY.md | ||
| uv.lock | ||
Caution
Users must update to Langflow >= 1.3 to protect against CVE-2025-3248.
Langflow is a powerful tool for building and deploying AI-powered agents and workflows. It provides developers with both a visual authoring experience and built-in API and MCP servers that turn every workflow into a tool 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 interface to quickly get started and iterate .
- Source code access lets you customize any component using Python.
- Interactive playground to immediately test and refine your flows with step-by-step control.
- Multi-agent orchestration with conversation management and retrieval.
- Deploy as an API or export as JSON for Python apps.
- Deploy as an MCP server and turn your flows into tools for MCP clients.
- Observability with LangSmith, LangFuse and other integrations.
- Enterprise-ready security and scalability.
⚡️ Quickstart
Langflow requires Python 3.10 to 3.13 and uv.
- To install Langflow, run:
uv pip install langflow -U
- To run Langflow, run:
uv run langflow run
- Go to the default Langflow URL at
http://127.0.0.1:7860.
For more information about installing Langflow, including Docker and Desktop options, see Install Langflow.
📦 Deployment
Langflow is completely open source and you can deploy it to all major deployment clouds. To learn how to use Docker to deploy Langflow, see the Docker deployment guide.
⭐ 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.