skills.rest
RADHA0-max
RADHA0-maxCOMMUNITY
Explore all skills by this author →
GitHub

add-atomic-chat-tool

Registers an MCP server exposing local Atomic Chat models as container agent tools.

2|Updated Jan 14, 2026
Search Tags:#typescript#docker#mcp server#agent tools#local models#atomic chat#openai-compatible api
One-click install
npx skills add https://github.com/RADHA0-max/selfhealbackend --skill add-atomic-chat-tool-radha0-max
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: add-atomic-chat-tool
Source: https://github.com/RADHA0-max/selfhealbackend/tree/main/nanoclaw/.claude/skills/add-atomic-chat-tool
Command: npx skills add https://github.com/RADHA0-max/selfhealbackend --skill add-atomic-chat-tool-radha0-max

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @modelcontextprotocol/sdk, zod.

What problem does it solve? Container agents in NanoClaw cannot call local models running in the Atomic Chat desktop app, forcing all work through the primary LLM even when a cheaper local model would suffice. ## Core Features & Use Cases - MCP Server Registration: Copies a stdio-based MCP server into the agent-runner tree and registers it in index.ts so the agent gains atomic_chat_list_models and atomic_chat_generate tools. - Env Forwarding: Adds a host-side helper that passes ATOMIC_CHAT_HOST and ATOMIC_CHAT_API_KEY into the container as Docker -e arguments. - Wiring Tests: Ships structural tests that verify the registration entry and the buildContainerArgs call actually exist, catching drift when edits are removed. - Use Case: Ask the agent to "use atomic chat to summarize this document" and it lists available local models, then generates a response via Atomic Chat's OpenAI-compatible API on port 1337. ## Quick Start Ask the agent to add the Atomic Chat tool so it can call local models, then verify by sending "use atomic chat to tell me the capital of France".

Frequently Asked Questions about add-atomic-chat-tool

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

FAQPage Schema
How do I add local Atomic Chat models as tools for a container agent?▼

Copy the MCP server source into the agent-runner tree, register an atomic_chat entry in the mcpServers object in index.ts, and forward the ATOMIC_CHAT_* env vars in container-runner.ts. Rebuild the container and restart the service to expose the tools.

What tools does the Atomic Chat MCP server expose?▼

It exposes atomic_chat_list_models, which queries GET /v1/models, and atomic_chat_generate, which sends prompts via POST /v1/chat/completions. Both call the Atomic Chat desktop app's OpenAI-compatible API.

Does Atomic Chat require an API key for local use?▼

No, Atomic Chat does not require authentication when running locally, so ATOMIC_CHAT_API_KEY can be left unset. Only set it if the app sits behind a reverse proxy that enforces auth.

Why does the agent say Atomic Chat is not installed?▼

This means the MCP server file was not copied, the atomic_chat entry is missing from the mcpServers object in index.ts, or the container image was not rebuilt. The tool allow-pattern is derived from the registered server name, so registration is the key check.

Why is the first Atomic Chat generation request slow?▼

Atomic Chat lazy-loads models into memory on first use, so the initial call takes longer while the model warms up. Subsequent calls against the same model are fast.

Related Skills

Explore More Software Engineering Skills→
wrightchinwrightchin

add-atomic-chat-tool

Expose local Atomic Chat models as MCP tools for NanoClaw agents.

Community
Advanced
tinner-deinnotinner-deinno

model-GPT

Sends prompts to the OpenAI API with automatic model selection and token logging.

Community
Intermediate
BartunekBartunek

add-ollama-tool

Add an Ollama MCP server to NanoClaw for local model calls.

Community
Intermediate
UniClipboardUniClipboard
1.6k

ui-iteration

Applies pixel-level UI adjustments one property at a time with screenshot verification.

Official
Intermediate