Community-maintained Model Context Protocol bridge that exposes Ollama's local HTTP API—model listing, pulls, chat, and OpenAI-compatible completions—to MCP clients such as Claude Desktop and Cursor. Published on npm as `ollama-mcp-server` (maintained fork of NightTrek/Ollama-mcp); requires a running Ollama daemon reachable at `OLLAMA_HOST` (default `http://127.0.0.1:11434`).
Use cases
- Run coding agents entirely against local models without cloud inference
- Pull and manage Ollama models from within an MCP-enabled assistant
- Prototype multimodal prompts when vision-capable models are installed locally
- Swap between cloud and local models during development with the same MCP surface
- Keep sensitive prompts on hardware you control while still using agent UX
Key features
- Claude Desktop
- Cursor
- Codex
Frequently Asked Questions
- Is this an official Ollama project?
- No—the npm package and repository are community-maintained; they wrap Ollama's documented HTTP API. Always review upstream release notes before upgrading.
- What prerequisites are required?
- Install Ollama from ollama.com and ensure the daemon is running. Configure `OLLAMA_HOST` if you use a non-default address or remote endpoint.
- Does chat streaming work over MCP stdio?
- The upstream documentation notes that in stdio mode responses return after completion rather than token streaming; plan UX accordingly.
Related
Related
3 Indexed items
Mem0 MCP Server
Mem0 documents a hosted Model Context Protocol server at https://mcp.mem0.ai/mcp that exposes Platform memory tools (`add_memory`, `search_memories`, `get_memories`, `update_memory`, `delete_memory`, `delete_all_memories`, `delete_entities`, `list_entities`, `list_events`, `get_event_status`) to Claude, Claude Code, Codex, Cursor, Windsurf, VS Code, and OpenCode. Setup uses `npx mcp-add` with HTTP transport or manual JSON/TOML client configs; Codex requires `MEM0_API_KEY` as bearer token per docs.mem0.ai/platform/mem0-mcp. The cloud server needs a Mem0 Platform API key from the dashboard and Node.js for the installer—no local vector database required for the hosted path.
Agent Protocol MCP
Implements the Agent Protocol standard enabling MCP clients to coordinate with external agent frameworks using shared task, step, and artifact schemas. Useful when composing multiple agents where one agent hands off work to specialized agents. Supports agent registration and state tracking.
Algolia Productivity MCP Server
Algolia documents an official managed Model Context Protocol server at algolia.com/doc/guides/model-context-protocol/productivity-mcp. Connect MCP clients to the remote HTTP endpoint `https://mcp.algolia.com/mcp` with OAuth (enable under Generate AI in the Algolia dashboard; sign in when prompted so the MCP inherits your account permissions). Productivity MCP is user-scoped and read-only per docs—tools cover search (`algolia_search_list_indices`, `algolia_search_index`, `algolia_search_for_facet_values`), Recommend (`algolia_recommendations`), and analytics helpers such as top searches, no-click rates, filter usage, and user counts. Algolia docs distinguish this from Algolia Public MCP for application-scoped, curated index exposure to external agents. Supported clients include ChatGPT, Claude, Claude Code, Cursor, Gemini CLI, VS Code, and OpenAI Playground.