open-webui

openapi-servers: Build OpenAPI Tool Servers for AI Agents

openapi-servers provides reference FastAPI implementations for exposing external tools and data through standard OpenAPI APIs. It suits developers integrating REST services with LLM agents who want an alternative to a proprietary tool protocol.

InactivePythonMITAI AgentsAPIMCP
Stars
1.1k
Forks
179
Stars, last 30 days
+3
Last commit
1.1 years ago
Contributors
7

Our take

Inactiveno commits in over a year

Health score

27/100

How it is scored
Activity
19
Community
46
Issues
21
Pull requests
20

+3 stars in the last 30 days

This is a practical set of OpenAPI-based starting points, but its inactive maintenance makes it a weaker choice when you need actively maintained reference implementations.

Good fit if

  • You want to build or adapt Python/FastAPI tool endpoints for OpenAPI-compatible AI clients.
  • You need examples spanning common integrations such as filesystem access, SQL, or RAG.
  • You are comfortable reviewing and maintaining the chosen server implementation yourself.

Look elsewhere if

  • You need a project with recent commits, releases, or evidence of ongoing maintenance.
  • You need a complete, supported deployment product rather than reference implementations to adapt.
  • You require a different primary interface and do not want to use an MCP bridge or another adapter.
All health signals
Last commit2025-09-25 (1.1 years ago)
Commits, last 90 days0
Releases, last 12 months0
Contributors7 (top contributor: 58% of commits)
Issues closed, last 90 days0
Pull requests merged, last 90 days0
Project age20 months

Checked on 2026-11-03 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

1,084

Forks

178

Change since Nov 17, 2026: +1 stars·0 forks

Overview

openapi-servers is a collection of reference tool servers built with FastAPI and the OpenAPI specification. It gives developers a standard HTTP/REST way to connect external tools and data sources to LLM agents and workflows, without requiring clients to use a project-specific protocol.

The examples cover file operations, Git, memory and knowledge graphs, weather, user information, SQL, and external RAG pipelines. They are intended as starting points to run or adapt, rather than as a single hosted product. OpenAPI-compatible clients can call the servers directly; optional bridges support integrations with MCP environments.

Key Features

  • Provides separate FastAPI server examples under servers/.
  • Demonstrates filesystem operations with configurable restrictions.
  • Includes examples for Git repository access and persistent memory or knowledge-graph queries.
  • Shows integrations for weather, user information, SQL chat, and external RAG pipelines.
  • Uses OpenAPI-described HTTP APIs that can be deployed locally or externally.
  • Documents optional MCP-to-OpenAPI bridging with mcpo or the included Python proxy.
  • Points to community-maintained options for exposing OpenAPI endpoints through MCP.

Use Cases

  • AI application developers can use the examples as a starting point for giving agents access to APIs and data through a familiar REST interface.
  • Teams with existing OpenAPI services can adapt the patterns to make those services callable by OpenAPI-compatible agent clients.
  • Developers migrating MCP tools can explore the documented MCP-to-OpenAPI bridge instead of rewriting every tool immediately.
  • Prototype builders can run a focused example, such as weather or filesystem access, locally before adapting it to their own environment.

What you need

Detected in the repository

  • Python (main language on GitHub)
  • A Compose file to run it with its related services

License in plain words

MITpermissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the copyright and license notice
  • Share your changes: no

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

Clone the repository and start the filesystem example:

git clone https://github.com/open-webui/openapi-servers
cd openapi-servers/servers/filesystem
pip install -r requirements.txt
uvicorn main:app --host 0.0.0.0 --reload

See the README for Docker Compose instructions, other server examples, and MCP bridge options.

Alternatives

  • mcpo: mcpo exposes existing MCP servers as REST APIs with generated OpenAPI schemas, while openapi-servers provides reference FastAPI implementations directly.
ProjectLanguageLicenseStarsStatus
openapi-serversPythonMIT1.1kInactive
mcpoPythonMIT4.4kQuiet

Considerations

The repository is marked inactive, with no commits in over a year and no releases in over a year. Treat the code as reference material and check each server's implementation and dependencies before exposing it to real users or sensitive data. The README describes the filesystem example as having configurable restrictions, but teams should verify that its controls meet their own security requirements. The repository also says further examples will be developed, so the set of available implementations may not cover every integration a team needs.

Found this useful?

Share it with someone who would like openapi-servers.

Comparisons