Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

Add Atomic Chat Tool

An installer skill that wires the NanoClaw container agent to local models served by the Atomic Chat desktop app via MCP.

Dev & CodingAdvanced★ 30,471⑂ 12,865AI score 7/10Last updated: Aug 9, 2026

What it does

Exposes models running in the local Atomic Chat desktop app (a Jan fork with an OpenAI-compatible API at http://127.0.0.1:1337/v1) as MCP tools for the NanoClaw container agent. The skill copies its bundled MCP server source and tests into both the container (Bun) tree and the host (Node) tree, registers an atomic_chat entry in the mcpServers map inside index.ts, and patches container-runner.ts to forward ATOMIC_CHAT_HOST / ATOMIC_CHAT_API_KEY into the container.

Two tools become available:

  • atomic_chat_list_models — list models currently loaded in Atomic Chat
  • atomic_chat_generate — send a prompt to a chosen model and return the reply

It also raises [ATOMIC] log lines to info level, appends env stubs to .env.example, and gives an explicit build / type-check / wiring-test validation sequence.

Who it's for

  • NanoClaw operators who want Claude to stay the orchestrator while offloading bulk work to local models
  • Anyone cutting API spend or keeping sensitive text on-device
  • Developers who want a worked example of wiring an MCP stdio server into a real codebase

Examples

  1. Ask "add the Atomic Chat integration" → the skill pre-flights the local API, copies files, registers the server, and runs the wiring tests.
  2. After install: "use atomic chat to tell me the capital of France" → the agent lists models, then generates with a local one.
  3. "Summarize these 200 log lines with llama3.2-3b-instruct" → long-text summarization is delegated locally to save tokens.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .claude/skills/add-atomic-chat-tool folder from the GitHub repo nanocoai/nanoclaw into my ~/.claude/skills/add-atomic-chat-tool/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/nanocoai/nanoclaw.git && cp -r nanoclaw/.claude/skills/add-atomic-chat-tool ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. On macOS, install Atomic Chat and launch the app.
  2. Open Settings → Local API Server, enable it on port 1337, then download a model from the Hub tab and send one message to warm it up.
  3. Confirm the API responds: curl -s http://127.0.0.1:1337/v1/models.
  4. Get the NanoClaw project: git clone https://github.com/nanocoai/nanoclaw.git
  5. The skill already lives at .claude/skills/add-atomic-chat-tool; to use it elsewhere run cp -r nanoclaw/.claude/skills/add-atomic-chat-tool ~/.claude/skills/.
  6. Open Claude Code at the NanoClaw project root and ask: "use the add-atomic-chat-tool skill to apply the Atomic Chat integration".
  7. Run the validation commands the skill prescribes (pnpm run build, the wiring/registration tests, ./container/build.sh) and make sure they all pass.
  8. Restart the service, verify with a message like "use atomic chat ...", and inspect tail -f logs/nanoclaw.log | grep -i atomic if anything fails.
View source on GitHub ↗License: MIT