Rune

The simplest agent harness for Claude Code
No SDK. No boilerplate. Just one file per agent.

Prerequisites

• Node.js 18+
• Claude Code CLI installed and logged in — Rune uses Claude Code under the hood for all agent execution

npm install -g @anthropic-ai/claude-code
claude                                       # login if you haven't

Install

npm install -g openrune

30-Second Harness

rune new reviewer --role "Code reviewer, security focused"
rune run reviewer.rune "Review the latest commit"

That's it. You just built an agent harness.

Core Concepts

One file = one agent

rune new architect --role "Software architect"
rune new coder --role "Backend developer"
rune new reviewer --role "Code reviewer"

Each .rune file is just JSON — portable, shareable, version-controllable:

{
  "name": "reviewer",
  "role": "Code reviewer, security focused",
  "history": [],
  "memory": []
}

Headless execution

Run agents from the terminal. No GUI needed:

rune run reviewer.rune "Review the latest commit"

# Pipe input from other commands
git diff | rune run reviewer.rune "Review this diff"

# JSON output for scripting
rune run reviewer.rune "Check for security issues" --output json

Autonomous mode

With --auto, agents write files, run commands, and fix errors on their own:

rune run coder.rune "Create an Express server with a /health endpoint. Run npm init and npm install." --auto

🔮 [auto] coder is working on: Create an Express server...

  ▶ Write: /path/to/server.js
  ▶ Bash: npm init -y
  ▶ Bash: npm install express
  💬 Server created and dependencies installed.

✓ coder finished

Agent pipeline

Chain agents. Each agent's output feeds into the next:

rune pipe architect.rune coder.rune "Build a REST API with Express"

With --auto, the last agent executes the plan:

rune pipe architect.rune coder.rune "Build a REST API with Express" --auto

architect designs → coder implements (writes files, installs deps).

Automated triggers

# On every git commit
rune watch reviewer.rune --on git-commit --prompt "Review this commit"

# On file changes
rune watch linter.rune --on file-change --glob "src/**/*.ts" --prompt "Check for issues"

# On a schedule
rune watch monitor.rune --on cron --interval 5m --prompt "Check server health"

Node.js API

Use agents in your own code. Each .send() call spawns a Claude Code process, so Claude Code CLI must be installed and logged in on the machine.

const rune = require('openrune')

const reviewer = rune.load('reviewer.rune')
const result = await reviewer.send('Review the latest commit')

// Pipeline
const { finalOutput } = await rune.pipe(
  ['architect.rune', 'coder.rune'],
  'Build a REST API'
)

Example: Agent-Powered API Server

A full walkthrough — from zero to a running server built by agents.

1. Install and create agents

npm install -g openrune
mkdir my-project && cd my-project

rune new architect --role "Software architect. Design system architecture concisely."
rune new coder --role "Backend developer. Implement code based on the given plan."
rune new reviewer --role "Code reviewer. Review for bugs and security issues."

2. Agents collaborate to build a server

rune pipe architect.rune coder.rune "Design and build an Express server with POST /review endpoint that uses require('openrune') to load reviewer.rune and send the prompt. Run npm init -y and npm install express openrune." --auto

architect designs the architecture → coder writes server.js, runs npm init, installs dependencies.

3. Start the server

node server.js

4. Call the agent via API

curl -X POST http://localhost:3000/review \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Review this project"}'

The reviewer agent analyzes your project and returns a full code review.

5. Open the desktop UI

rune open reviewer.rune

The conversation history from the API call is already there — context persists across CLI, API, and GUI.

Desktop UI

Rune includes an optional desktop app for interactive chat.

rune open reviewer.rune

Or double-click any .rune file in Finder.

• Real-time activity — See tool calls, results, and permission requests as they happen.
• Built-in terminal — Claude Code output and your own commands, side by side.
• Right-click to create — macOS Quick Action for creating agents from Finder.

If double-click doesn't work, run rune install once to register the file association.

CLI Reference

Watch events: git-commit, git-push, file-change (with --glob), cron (with --interval)

Platform Support

License

MIT