ollama-mcp-host is a translator that lets a local Ollama model call multiple MCP servers simultaneously, enabling tool-augmented AI entirely on your own machine.

Stars

22

7-day growth

No data

Forks

5

Open issues

0

License

MIT

Last updated

2026-06-29

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It enables air-gapped, privacy-preserving AI by bridging Ollama and MCP without cloud dependencies, with a tiny footprint (single dependency) and built-in handling of Gemma's thinking mode to ensure reliable tool calls.

Who it is for

  • Developers building local AI assistants with tool integration
  • Privacy-conscious users requiring no-cloud AI inference
  • Organizations needing air-gapped or on-premise AI solutions
  • Users of Korean public-sector MCP servers (law search and document generation)

Use cases

  • Running a local AI that can search legal statutes and generate official documents via MCP tools
  • Deploying a private AI assistant inside an offline network without internet access
  • Prototyping multi-tool AI workflows using a single Ollama model
  • Exposing an HTTP API for other applications to query a tool-capable local model

Strengths

  • 100% local inference; no cloud or API keys required for the LLM
  • Single runtime dependency (@modelcontextprotocol/sdk) – minimal overhead
  • Supports both interactive CLI and programmatic library usage
  • Automatically disables Gemma's thinking mode to ensure tool calls fire reliably

Considerations

  • Requires Node.js 20+, Ollama with a tool-capable model, and MCP servers configured via stdio
  • Offline setup requires manual mirroring of model files, npm packages, and source code
  • Firewall or network restrictions may block MCP server outbound calls (e.g., to law.go.kr)

README quick start

Quick start

Description

로컬 Ollama(gemma 등)로 korean-law-mcp·kordoc 같은 여러 MCP 서버를 동시에 돌리는 호스트. 클라우드 API 없이 내부망/오프라인 동작.

Related repositories

Similar projects matched by category, topics, and programming language.

TanStack
Featured
TanStack GitHub avatar

router

TanStack Router is a type-safe, data-driven React router with built-in caching, prefetching, and nested layouts, while TanStack Start extends it into a full-stack SSR framework.

Web DevelopmentFrontend Frameworks
14,861
vercel-labs
Featured
vercel-labs GitHub avatar

scriptc

scriptc compiles ordinary TypeScript into small, fast native executables without needing Node.js, V8, or any JavaScript runtime in the binary.

Developer ToolsCode Quality & Build
1,985
Jakubantalik
Featured
Jakubantalik GitHub avatar

thinking-orbs

A React component library that renders six hand-tuned animated thought orb loading indicators on a plain 2D canvas, with two purpose-tuned sizes and automatic theme detection for AI and agent UIs.

AI & Machine LearningAI Agents
1,191