Skip to main content
Glama
Sign In
ollama-mcp

ollama-mcp

by DTG404
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Local

ollama-mcp

VRAM-aware MCP server that offloads token-heavy development tasks to local Ollama models, saving API tokens for complex reasoning.

Tools

Tool

Description

draft_code

Generate boilerplate and scaffolding

draft_commit_message

Write commit messages from diffs

draft_content

Draft documentation and prose

classify_task

Auto-route work to the right agent type

summarize_file

Summarize unfamiliar source files

triage_issues

Classify and prioritize issues

list_models

List available Ollama models

check_updates

Check for model updates

get_config

Show current configuration

Related MCP server: token-diet

VRAM Management

The server monitors GPU memory via nvidia-smi and selects models based on available VRAM. Models are loaded on demand.

Setup

npm install

Requires Ollama running locally with models pulled.

Usage

npm start

Configure in Claude Code MCP settings:

{
  "mcpServers": {
    "ollama": {
      "command": "node",
      "args": ["/path/to/ollama-mcp/index.js"]
    }
  }
}

Tech Stack

  • Node.js (>=18)

  • @modelcontextprotocol/sdk

  • Zod for schema validation

  • Ollama API for local model inference

  • nvidia-smi for VRAM monitoring

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    This server cannot be deployed

    Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers

    • knitbrain

      PDgit12
      A
      license
      A
      quality
      C
      maintenance
      Local-first MCP server that gives any AI coding agent per-project memory, workflow intelligence, and always-on, lossless token & context optimization.
      37
      65 npm
      6
      MIT

    • token-diet

      aryxnsdfs
      A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that reduces token usage by injecting graph-ranked repo maps, decision logs, and diff-only output into AI coding tool requests.
      2
      MIT

    • ai-mind-map

      shdra06
      A
      license
      Not graded
      quality
      C
      maintenance
      An MCP server that reduces AI coding agent token usage by 80-99% through a queryable knowledge graph of code, change tracking, and persistent memory across sessions.
      94 npm
      4
      MIT