open-webui-mcp
Exposes OpenWebUI's admin REST API as an MCP server, enabling administrative operations on OpenWebUI through natural language via MCP tools.
README
open-webui-mcp
A vendored fork of stephanschielke/open-webui-mcp-server packaged for container deployment, with a SHA-pinned upstream snapshot, a published ghcr.io image, and a small drift-detection helper.
Vendored at upstream commit 85d88af6e3dd34183a5a0eefb85d474f3e3c2b54 (2026-04-07). The wrapper source under src/openwebui_mcp/ is byte-identical to upstream — see NOTICE for full attribution.
What it does
Exposes OpenWebUI's admin REST API as a Model Context Protocol (MCP) server. Tool definitions are generated from a bundled snapshot of OpenWebUI's OpenAPI spec via FastMCP; 317 tools survive the upstream RouteMap filter (after excluding /ollama/*, /openai/*, /api/v1/{analytics,evaluations,terminals,pipelines}/*, and any non-/api/v1/ paths).
This image is intended to run behind an authenticating MCP gateway (e.g. LiteLLM, an mTLS proxy, or anything that fronts streamable-HTTP MCP). It exposes the full OpenWebUI admin surface — including mutating and destructive operations — by design. Do not expose port 7999 to untrusted networks; always front it with bearer auth or equivalent.
Deployment
services:
open-webui-mcp:
image: ghcr.io/tetra-2023/open-webui-mcp:stable
environment:
WEBUI_URL: http://open-webui:8080 # base URL of your OpenWebUI
WEBUI_API_KEY: ${WEBUI_API_KEY} # admin token sourced from your secrets store
MCP_TRANSPORT: http
MCP_HTTP_HOST: 0.0.0.0
MCP_HTTP_PORT: "7999"
MCP_HTTP_PATH: /mcp
depends_on: [open-webui]
Pin to a specific tag (:0.2.2) or digest in production. :stable floats with the most recent release tag; :latest floats with main.
Local run (debug)
docker run --rm -p 7999:7999 \
-e WEBUI_URL=http://host.docker.internal:8080 \
-e WEBUI_API_KEY=$WEBUI_API_KEY \
-e MCP_TRANSPORT=http \
-e MCP_HTTP_HOST=0.0.0.0 \
ghcr.io/tetra-2023/open-webui-mcp:stable
Spec drift check
When upgrading OpenWebUI, run the drift check before bumping this wrapper. The bundled OpenAPI snapshot drives tool generation, so a mismatch between the snapshot and live OWUI causes the wrapper to advertise tool schemas that no longer match the running endpoints.
WEBUI_API_KEY=... ./scripts/check-spec-drift.sh https://owui.example.com
Reports added / removed / renamed operations between the bundled snapshot and the live OWUI's /openapi.json. See CONTRIBUTING.md for the full bump procedure.
Tags published on ghcr.io
| Tag | When | Source |
|---|---|---|
:latest |
every push to main |
CI docker job |
:stable |
every v*.*.* git tag |
Release Image workflow |
:<version> |
every v*.*.* git tag |
Release Image workflow |
License
MIT — see LICENSE (verbatim from upstream) and NOTICE (fork attribution + modifications).
Refs
- Upstream source: https://github.com/stephanschielke/open-webui-mcp-server
- Original fork: https://github.com/troylar/open-webui-mcp-server
- OpenWebUI upstream RFE for a first-party admin MCP: https://github.com/open-webui/open-webui/discussions/16891
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.