Langflow

aid: langflow
url: https://raw.githubusercontent.com/api-evangelist/langflow/refs/heads/main/apis.yml
apis:
  - aid: langflow:langflow-api
    name: Langflow API
    tags:
      - AI
      - Agents
      - Workflows
      - Low-Code
      - LangChain
    humanURL: https://docs.langflow.org/api-reference-api-examples
    baseURL: http://localhost:7860/api
    properties:
      - url: https://docs.langflow.org/api-reference-api-examples
        type: Documentation
      - url: https://docs.langflow.org/api-keys-and-authentication
        type: Authentication
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
      - url: json-ld/langflow-context.jsonld
        type: JSONLD
      - url: vocabulary/langflow-vocabulary.yml
        type: Vocabulary
      - url: rules/langflow-rules.yml
        type: SpectralRuleset
    description: >-
      Langflow's FastAPI-based REST API exposes every capability of a running Langflow server — running flows, building
      flows, managing projects, files, users, API keys, MCP servers, messages and sessions, traces, starter projects,
      and an OpenAI-compatible Responses endpoint. Authentication is via the `x-api-key` header or query parameter. An
      interactive OpenAPI 3.1 spec is served at `/docs` and `/openapi.json` on every deployment.
  - aid: langflow:langflow-flows-api
    name: Langflow Flows API
    tags:
      - AI
      - Flows
      - Agents
    humanURL: https://docs.langflow.org/concepts-flows
    properties:
      - url: https://docs.langflow.org/concepts-flows
        type: Documentation
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
      - url: json-schema/langflow-flow-schema.json
        type: JSONSchema
    description: >-
      Create, read, update, delete, upload, download, and list flows. A flow is the canonical Langflow unit — a directed
      graph of components representing an AI agent or workflow. Includes endpoints for batch creation, basic examples,
      public flows, and starter projects.
  - aid: langflow:langflow-run-api
    name: Langflow Run API
    tags:
      - AI
      - Agents
      - Execution
    humanURL: https://docs.langflow.org/api-reference-api-examples
    properties:
      - url: https://docs.langflow.org/api-reference-api-examples
        type: Documentation
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      Execute a flow by id or name via POST `/api/v1/run/{flow_id_or_name}` (simplified), POST
      `/api/v1/run/advanced/{flow_id_or_name}` (experimental advanced), or POST `/api/v1/webhook/{flow_id_or_name}` for
      webhook-triggered runs. Returns flow outputs and supports streaming via the `stream` query parameter.
  - aid: langflow:langflow-build-api
    name: Langflow Build API
    tags:
      - AI
      - Build
      - Streaming
    humanURL: https://docs.langflow.org/api-reference-api-examples
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      Build endpoints run a flow with full vertex-level streaming. POST `/api/v1/build/{flow_id}/flow` kicks off a build
      job and returns a job id; GET `/api/v1/build/{job_id}/events` streams vertex build events; POST
      `/api/v1/build/{job_id}/cancel` cancels in-flight builds. Public-tmp variants exist for unauthenticated public
      flow execution.
  - aid: langflow:langflow-projects-api
    name: Langflow Projects API
    tags:
      - AI
      - Projects
      - Organization
    humanURL: https://docs.langflow.org/concepts-flows-import
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
      - url: json-schema/langflow-project-schema.json
        type: JSONSchema
    description: >-
      Create, list, read, update, delete, upload, and download Langflow projects. Projects are containers that group
      flows and provide MCP server installation context.
  - aid: langflow:langflow-files-api
    name: Langflow Files API
    tags:
      - AI
      - Files
      - Storage
    humanURL: https://docs.langflow.org/concepts-file-management
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      Upload, download, list, and delete files attached to a flow (v1) or to the authenticated user (v2). The v2 Files
      API supports batch upload/download, batch delete, and profile-picture management.
  - aid: langflow:langflow-monitor-api
    name: Langflow Monitor API
    tags:
      - AI
      - Monitoring
      - Messages
      - Sessions
    humanURL: https://docs.langflow.org/concepts-flows-monitor
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
      - url: json-schema/langflow-message-schema.json
        type: JSONSchema
    description: >-
      Inspect and manage execution history — vertex builds, chat messages, sessions, and shared sessions. Supports
      listing, updating, renaming, and deleting messages and entire sessions, plus shared-session sharing for read-only
      collaboration.
  - aid: langflow:langflow-traces-api
    name: Langflow Traces API
    tags:
      - AI
      - Tracing
      - Observability
    humanURL: https://docs.langflow.org/concepts-flows-monitor
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      Retrieve and delete execution traces. GET `/api/v1/monitor/traces` lists traces with flow-id filtering; GET
      `/api/v1/monitor/traces/{trace_id}` returns a single trace. Integrates with external observability providers like
      LangSmith and LangFuse.
  - aid: langflow:langflow-users-api
    name: Langflow Users API
    tags:
      - AI
      - Users
      - Administration
    humanURL: https://docs.langflow.org/configuration-authentication
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
      - url: json-schema/langflow-user-schema.json
        type: JSONSchema
    description: >-
      User management endpoints — create, read, patch, and delete users; `/api/v1/users/whoami` returns the current
      authenticated user; reset-password endpoint for superuser-driven credential resets.
  - aid: langflow:langflow-mcp-api
    name: Langflow MCP API
    tags:
      - AI
      - MCP
      - Model Context Protocol
      - Servers
    humanURL: https://docs.langflow.org/mcp-server
    properties:
      - url: https://docs.langflow.org/mcp-server
        type: Documentation
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
      - url: json-schema/langflow-mcp-server-schema.json
        type: JSONSchema
    description: >-
      Manage MCP (Model Context Protocol) servers attached to Langflow projects and globally. Every Langflow project can
      be exposed as an MCP server so its flows are callable as MCP tools. Endpoints list, register, update, and remove
      MCP servers, and install MCP configurations into Claude Desktop and other MCP hosts.
  - aid: langflow:langflow-workflow-api
    name: Langflow Workflow API
    tags:
      - AI
      - Workflows
      - Execution
    humanURL: https://docs.langflow.org/concepts-flows
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      v2 Workflow execution API — POST `/api/v2/workflows` executes a workflow; GET returns workflow status; POST
      `/api/v2/workflows/stop` halts a running workflow. The v2 workflow surface is the forward-looking execution
      interface for Langflow agents.
  - aid: langflow:langflow-openai-responses-api
    name: Langflow OpenAI Responses API
    tags:
      - AI
      - OpenAI-Compatible
      - Responses
    humanURL: https://docs.langflow.org/api-reference-api-examples
    properties:
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      OpenAI-compatible Responses endpoint at POST `/api/v1/responses`. Lets clients written against OpenAI's Responses
      API point at a Langflow flow without code changes — the flow plays the role of the model and tool-calling layer.
  - aid: langflow:langflow-api-keys-api
    name: Langflow API Keys API
    tags:
      - AI
      - Authentication
      - Administration
    humanURL: https://docs.langflow.org/api-keys-and-authentication
    properties:
      - url: https://docs.langflow.org/api-keys-and-authentication
        type: Documentation
      - url: openapi/langflow-openapi.yml
        type: OpenAPI
    description: >-
      Create, list, rotate, and delete Langflow API keys. Keys authenticate requests against any of the Langflow REST
      endpoints via the `x-api-key` header. Also creatable via the `langflow api-key` CLI when the frontend is disabled.
name: Langflow
tags:
  - AI
  - Artificial Intelligence
  - Agents
  - Workflows
  - Low-Code
  - Visual Builder
  - LangChain
  - RAG
  - MCP
  - Open Source
  - FastAPI
kind: contract
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
access: 3rd-Party
common:
  - url: https://www.langflow.org
    type: Portal
  - url: https://docs.langflow.org
    name: Langflow Documentation
    type: Documentation
  - url: https://docs.langflow.org/get-started-installation
    name: Installation
    type: GettingStarted
  - url: https://docs.langflow.org/get-started-quickstart
    name: Quickstart
    type: GettingStarted
  - url: https://docs.langflow.org/api-reference-api-examples
    name: API Reference
    type: Documentation
  - url: https://docs.langflow.org/api-keys-and-authentication
    name: API Keys and Authentication
    type: Authentication
  - url: https://docs.langflow.org/concepts-flows
    name: Flows Concepts
    type: Documentation
  - url: https://docs.langflow.org/concepts-components
    name: Components Reference
    type: Documentation
  - url: https://docs.langflow.org/configuration-authentication
    name: Authentication Configuration
    type: Documentation
  - url: https://docs.langflow.org/deployment-overview
    name: Deployment Overview
    type: Documentation
  - url: https://docs.langflow.org/deployment-docker
    name: Docker Deployment
    type: Documentation
  - url: https://docs.langflow.org/deployment-kubernetes
    name: Kubernetes Deployment
    type: Documentation
  - url: https://docs.langflow.org/release-notes
    name: Release Notes
    type: ChangeLog
  - url: https://github.com/langflow-ai/langflow
    name: Langflow GitHub
    type: SourceCode
  - url: https://github.com/langflow-ai
    name: Langflow GitHub Organization
    type: GitHubOrganization
  - url: https://github.com/langflow-ai/langflow/blob/main/LICENSE
    name: MIT License
    type: License
  - url: https://github.com/langflow-ai/langflow/releases
    name: Releases
    type: ChangeLog
  - url: https://github.com/langflow-ai/langflow/issues
    name: Issues
    type: IssueTracker
  - url: https://github.com/langflow-ai/langflow/blob/main/CONTRIBUTING.md
    name: Contributing
    type: Documentation
  - url: https://github.com/langflow-ai/langflow-helm-charts
    name: Langflow Helm Charts
    type: HelmChart
  - url: https://github.com/langflow-ai/langflow-embedded-chat
    name: Langflow Embedded Chat
    type: Tool
  - url: https://github.com/langflow-ai/langflow-client-ts
    name: TypeScript Client
    type: SDK
  - url: https://github.com/langflow-ai/langflow-railway
    name: Railway Template
    type: Deployment
  - url: https://github.com/langflow-ai/openrag
    name: OpenRAG
    type: Tool
  - url: https://github.com/langflow-ai/langflow-bundles
    name: Langflow Bundles
    type: Tool
  - url: https://github.com/langflow-ai/langflow-twilio-voice
    name: Twilio Voice Example
    type: CodeExamples
  - url: https://github.com/langflow-ai/mcp-sse-shim
    name: MCP SSE Shim
    type: Tool
  - url: https://pypi.org/project/langflow/
    name: PyPI Package
    type: Package
  - url: https://hub.docker.com/r/langflowai/langflow
    name: Docker Image
    type: Container
  - url: https://docs.langflow.org/develop-application
    name: Develop with Langflow
    type: Documentation
  - url: https://docs.langflow.org/concepts-flows-import
    name: Import and Export Flows
    type: Documentation
  - url: https://docs.langflow.org/concepts-file-management
    name: File Management
    type: Documentation
  - url: https://docs.langflow.org/concepts-flows-monitor
    name: Monitor Flows
    type: Documentation
  - url: https://docs.langflow.org/concepts-publish
    name: Publish Flows
    type: Documentation
  - url: https://docs.langflow.org/concepts-playground
    name: Playground
    type: Documentation
  - url: https://docs.langflow.org/agents-overview
    name: Agents Overview
    type: Documentation
  - url: https://docs.langflow.org/mcp-server
    name: MCP Server
    type: Documentation
  - url: https://docs.langflow.org/mcp-client
    name: MCP Client
    type: Documentation
  - url: https://docs.langflow.org/typescript-client
    name: TypeScript Client Docs
    type: Documentation
  - url: https://discord.com/invite/EqksyE2EX9
    name: Discord Community
    type: Forum
  - url: https://www.youtube.com/@Langflow
    name: YouTube Channel
    type: Video
  - url: https://twitter.com/langflow_ai
    name: Twitter
    type: Twitter
  - url: https://www.linkedin.com/company/langflow-ai/
    name: LinkedIn
    type: LinkedIn
  - url: https://www.langflow.org/blog
    type: Blog
  - url: https://www.langflow.org/use-cases
    type: UseCases
  - url: https://www.langflow.org/newsletter
    type: Newsletter
  - url: https://www.ibm.com/products/datastax
    name: IBM DataStax (Owner)
    type: Owner
  - type: Features
    data:
      - Visual drag-and-drop builder for AI agents, RAG pipelines, and LangChain workflows
      - FastAPI-based REST API with OpenAPI 3.1 spec served at /docs and /openapi.json on every deployment
      - 67 REST endpoints covering flows, builds, projects, files, users, API keys, MCP servers, monitoring, and traces
      - OpenAI-compatible Responses endpoint (/api/v1/responses) so OpenAI clients can target a Langflow flow
      - Webhook execution endpoint per flow for event-driven invocation
      - Streaming flow execution via SSE on the build endpoints
      - Native MCP (Model Context Protocol) server — every Langflow project is exposable as an MCP server
      - MCP client support for consuming external MCP servers as Langflow components
      - Project / flow / component hierarchy with import-export, batch operations, and public-flow sharing
      - Session-aware chat with shared sessions for read-only collaboration
      - Built-in trace explorer plus integrations with LangSmith and LangFuse for observability
      - Pluggable Python components — every component's source is editable in the UI
      - Multi-agent orchestration with conditional routing and tool calls
      - Vector-store integrations including Astra DB, Chroma, Pinecone, Milvus, Weaviate, Qdrant, and pgvector
      - >-
        LLM integrations including OpenAI, Anthropic, Google, Azure, Bedrock, Mistral, Cohere, Hugging Face, Ollama, and
        Groq
      - File upload and per-user file management with batch operations (v2 Files API)
      - API key authentication via `x-api-key` header or query parameter
      - Auto-login mode for local dev and superuser mode for production
      - >-
        Distributed by Python package on PyPI (`pip install langflow`), Docker image (`langflowai/langflow:latest`),
        Helm chart, and Desktop app for macOS and Windows
      - MIT-licensed, written in Python (FastAPI backend) and TypeScript (React Flow frontend)
      - 149k+ GitHub stars, v1.9.3 (May 2026) — actively maintained by langflow-ai with 800+ contributors
      - Hosted Langflow Cloud offering operated by IBM DataStax (post-acquisition)
    sources:
      - https://www.langflow.org
      - https://docs.langflow.org
      - https://github.com/langflow-ai/langflow
      - https://docs.langflow.org/api-reference-api-examples
      - https://docs.langflow.org/api-keys-and-authentication
    updated: '2026-05-24'
created: '2026-05-24T00:00:00.000Z'
modified: '2026-05-24'
position: Producing
description: >-
  Langflow is an open-source low-code visual builder for AI agents, RAG pipelines, and LangChain-based workflows. It
  pairs a drag-and-drop React Flow frontend with a FastAPI backend that exposes every flow as a REST API, an MCP server,
  and an OpenAI-compatible Responses endpoint. Components are editable Python and ship with integrations across most
  major LLMs, vector stores, and observability platforms. Langflow was acquired by DataStax in 2025; DataStax itself was
  acquired by IBM and the deal closed on May 28, 2025, making Langflow an IBM property while remaining MIT-licensed open
  source. The project is the canonical reference implementation for visually composing LangChain agents — 149k+ GitHub
  stars, distributed via PyPI, Docker, Helm, and native Desktop apps, with a hosted cloud option run by DataStax.
maintainers:
  - FN: Kin Lane
    email: info@apievangelist.com
    X: apievangelist
    url: https://apievangelist.com
specificationVersion: '0.16'