🌟 OpenAPI Tool Servers

This repository provides reference OpenAPI Tool Server implementations making it easy and secure for developers to integrate external tooling and data sources into LLM agents and workflows. Designed for maximum ease of use and minimal learning curve, these implementations utilize the widely adopted and battle-tested OpenAPI specification as the standard protocol.

By leveraging OpenAPI, we eliminate the need for a proprietary or unfamiliar communication protocol, ensuring you can quickly and confidently build or integrate servers. This means less time spent figuring out custom interfaces and more time building powerful tools that enhance your AI applications.

☝️ Why OpenAPI?

• Established Standard: OpenAPI is a widely used, production-proven API standard backed by thousands of tools, companies, and communities.

• No Reinventing the Wheel: No additional documentation or proprietary spec confusion. If you build REST APIs or use OpenAPI today, you're already set.

• Easy Integration & Hosting: Deploy your tool servers externally or locally without vendor lock-in or complex configurations.

• Strong Security Focus: Built around HTTP/REST APIs, OpenAPI inherently supports widely used, secure communication methods including HTTPS and well-proven authentication standards (OAuth, JWT, API Keys).

• Future-Friendly & Stable: Unlike less mature or experimental protocols, OpenAPI promises reliability, stability, and long-term community support.

🚀 Quickstart

Get started quickly with our reference FastAPI-based implementations provided in the servers/ directory. (You can adapt these examples into your preferred stack as needed, such as using FastAPI, FastOpenAPI or any other OpenAPI-compatible library):

git clone https://github.com/open-webui/openapi-servers
cd openapi-servers

# Example: Installing dependencies for a specific server 'filesystem'
cd servers/filesystem
pip install -r requirements.txt
uvicorn main:app --host 0.0.0.0 --reload

Or using Docker:

cd servers/filesystem
docker compose up

Now, simply point your OpenAPI-compatible clients or AI agents to your local or publicly deployed URL—no configuration headaches, no complicated transports.

📂 Server Reference Implementations

Reference implementations provided in this repository demonstrate common use-cases clearly and simply:

• Filesystem Access - Manage local file operations safely with configurable restrictions.
• Git Server - Expose Git repositories for searching, reading, and possibly writing via controlled API endpoints.
• Memory & Knowledge Graph - Persistent memory management and semantic knowledge querying using popular and reliable storage techniques.
• Weather Server - Provide current weather conditions and forecasts from trusted public APIs.
• Get User Info Server - Access and return enriched user profile information from authentication providers or internal systems.
• WIP: Web Search & Fetch - Retrieve and convert web-based content securely into structured API results usable by LLMs.
• WIP: Database Server - Query and inspect database schemas across common DB engines like PostgreSQL, MySQL, and SQLite.

(More examples and reference implementations will be actively developed and continually updated.)

[!IMPORTANT]
💡 Contribute Your Server!

We strongly encourage the community to contribute their own OpenAPI tool server examples! This is more important than it might seem: The world doesn’t need another closed protocol or proprietary format gatekeeping innovation—we need clearly defined, open, and composable APIs backed by open documentation and proven tools. OpenAPI is the future-proof foundation we can all build on—together.

Let’s build an open ecosystem where every tool speaks the same language—yours.

🔌 Bridge MCP → OpenAPI (Optional)

For the easiest way to expose your MCP tools as OpenAPI-compatible APIs, we recommend using mcpo. This enables tool providers who initially implemented MCP servers to expose them effortlessly as standard OpenAPI-compatible APIs, ensuring existing MCP servers and resources remain accessible without additional hassle.

Quick Usage:

uvx mcpo --port 8000 -- uvx mcp-server-time --local-timezone=America/New_York

Alternatively, we also provide a simple Python-based proxy server:

Example:

cd servers/mcp-proxy
pip install -r requirements.txt
python main.py --host 0.0.0.0 --port 8000 -- uvx mcp-server-time --local-timezone=America/New_York

Both methods help bridge existing MCP servers with OpenAPI clients, removing transport and security complexities during integration or migration.

🔃 Bridge OpenAPI → MCP (Optional)

In addition to turning MCP tool servers into OpenAPI-compatible APIs (via mcpo), it's also possible to go the other direction: expose your OpenAPI-compatible tool server as an MCP server. This makes it easy to integrate any OpenAPI tool into environments or agents that expect an MCP-compatible interface.

Several community-maintained projects are available to help with this:

• 🌉 openapi-mcp-server
  Acts as a translator from any OpenAPI spec to an MCP tool, allowing for easy reuse of RESTful APIs within MCP-only frameworks.

• 🔁 mcp-openapi-server
  A lightweight adapter that converts OpenAPI-described endpoints to usable MCP tool servers on the fly.

• 🌀 mcp-openapi-proxy
  Wraps OpenAPI endpoints in a proxy that re-expresses them in MCP-compatible format with minimal configuration.

• ⚡ fastapi_mcp
  A FastAPI extension that allows you to serve native FastAPI endpoints directly through the MCP protocol.

With these bridges, you gain full interoperability in both directions:

• Run OpenAPI tools using existing MCP pipelines
• Upgrade or replace MCP-native tools with more stable, secure, OpenAPI-based implementations

This two-way bridge is a major step forward in simplifying tool architecture, empowering developers to choose the most efficient and future-proof infrastructure without compromising compatibility.

📜 License

Licensed under MIT License.

🌱 Open WebUI Community

• For general discussions, technical exchange, and announcements, visit our Community Discussions page.
• Have ideas or feedback? Please open an issue!