skills.rest
jeffreykthomas
jeffreykthomasCOMMUNITY
Explore all skills by this author →
GitHub

add-ollama-tool

Expose on-host Ollama models to NanoClaw agents via an MCP server.

Updated Mar 5, 2026
Search Tags:#agent#mcp#container#inference#ollama#local-model#on-host
One-click install
npx skills add https://github.com/jeffreykthomas/nanoclaw-golf --skill add-ollama-tool
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: add-ollama-tool
Source: https://github.com/jeffreykthomas/nanoclaw-golf/tree/main/.claude/skills/add-ollama-tool
Command: npx skills add https://github.com/jeffreykthomas/nanoclaw-golf --skill add-ollama-tool

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Expose local Ollama models to NanoClaw agents through a dedicated MCP server, reducing latency by enabling on-host inference and avoiding remote calls.

Core Features & Use Cases

  • Exposes local Ollama models as MCP tools: ollama_list_models and ollama_generate.
  • Enables on-host inference for cost-effective tasks such as summarization and translation within container-driven workflows.
  • Supports per-group connectivity via the Ollama host configuration (OLLAMA_HOST) with sensible defaults.

Quick Start

Install Ollama on the host, ensure it is running, then apply this skill using the skills engine to enable the Ollama MCP server.

Frequently Asked Questions about add-ollama-tool

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I expose local Ollama models to container agents via MCP?▼

You can expose local Ollama models to container agents via MCP by applying this skill, which installs an MCP server and configures host connectivity to enable on-host inference for your agents.

What is the best way to run on-host inference for container-based agent tasks?▼

Running on-host inference for container-based tasks is best achieved by exposing local Ollama models through a dedicated MCP server, which reduces latency by avoiding remote API calls during generation.

Do I need Ollama installed on my host before applying this MCP integration?▼

Yes, you need to install Ollama on your host machine and ensure it is running before applying this skill to configure the MCP server and expose the ollama_generate and ollama_list_models tools.

Can I use local Ollama models for summarization and translation within container workflows?▼

Yes, you can use local Ollama models for cost-effective summarization and translation tasks by leveraging the on-host inference capabilities exposed through the MCP server integration.

How does the MCP server configure Ollama host connectivity for per-group agent integration?▼

The MCP server configures Ollama host connectivity using the OLLAMA_HOST environment variable, providing sensible defaults while allowing per-group agent integration within the container runtime.

What steps are involved in setting up the Ollama MCP server in a container runtime?▼

Setting up the Ollama MCP server involves a four-phase workflow: init, apply, configure, and verify, which sequentially installs, applies, and tests the local model integration within the container runtime.

Related Skills

Explore More Software Engineering Skills→
VincentChong123VincentChong123

add-ollama-tool

Integrate local Ollama models as MCP tools in containerized agents.

Community
Advanced
E01NE01N

add-ollama-tool

Expose local Ollama models through an MCP server with two tools.

Community
Advanced
theowenyoungtheowenyoung

add-ollama-tool

Expose local Ollama models to container agents via MCP tools.

Community
Advanced
VincentChong123VincentChong123

add-ollama-provider

Route NanoClaw agent traffic to a local Ollama instance via environment variables.

Community
Advanced