Reftrixmcp

Please install the `reftrixmcp` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Node.js 20+, pnpm 10+, Docker, Ollama** — Required runtimes. Ollama with llama3.2-vision model (~7.9GB) required for vision analysis. Run: `ollama pull llama3.2-vision` (https://ollama.com)
- **Setup infrastructure and build** — Clone, install, setup DB/Redis via Docker, build, install Playwright, download DINOv2 model Run: `git clone https://github.com/TKMD/ReftrixMCP.git && cd ReftrixMCP && pnpm install && cp .env.example .env.local && pnpm docker:up && pnpm db:migrate && pnpm db:seed && pnpm build && pnpm exec playwright install chromium && pnpm --filter @reftrixmcp/ml download:dinov2`

Canonical MCP server config (stdio transport):
- command: `node`
- args: ["/absolute/path/to/ReftrixMCP/apps/mcp-server/dist/index.js"]
- required environment variables:
  - `DATABASE_URL`: PostgreSQL connection string (example: `postgresql://reftrix:<password>@localhost:26432/reftrix?schema=public`)
  - `REDIS_URL`: Redis connection URL (example: `redis://localhost:27379`)
  - `OLLAMA_BASE_URL`: Ollama server URL for vision analysis (example: `http://localhost:11434`)
  - `OLLAMA_HOST`: Ollama host for worker process (example: `http://localhost:11434`)

Note: Web design knowledge base platform. 35 MCP tools for layout analysis, motion detection, quality scoring, and semantic search of UI patterns. Requires significant setup: PostgreSQL, Redis, Ollama + llama3.2-vision, DINOv2 (~800MB), Playwright. Min 16GB RAM.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.