A local-first filesystem message bus and webhook relay that enables seamless real-time coordination between multiple AI coding agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ide-agent-kit
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install ide-agent-kit using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is IDE Agent Kit?
IDE Agent Kit is a developer-centric toolkit designed to bridge the gap between isolated AI agents, turning them into a cohesive team. By utilizing a filesystem-based message bus, it allows agents to share events, poll rooms, and coordinate tasks across different sessions without mandatory network dependencies. It is a foundational component for those building advanced Openclaw Skills, providing the infrastructure needed for multi-model agent coordination, automated task execution, and real-time event handling.
The tool is built with a security-first mindset, operating locally by default. It ensures that sensitive data stays within your working directory while providing opt-in paths for advanced connectivity, such as GitHub webhook ingestion or Ant Farm room polling. This makes it an essential utility for developers who need to scale their agentic workflows while maintaining strict control over their environment.
IDE Agent Kit Use Cases
- Orchestrating multiple AI agents to work together on complex coding tasks across different IDE windows.
- Automatically triggering agent workflows based on inbound GitHub or GitLab webhooks.
- Sharing persistent memory and state between different agent sessions using a local backend.
- Running scheduled maintenance jobs or automated test suites through integrated cron support.
- Executing shell commands through a secure, allowlisted, and auditable approval workflow.
How IDE Agent Kit Works
- The user initializes the kit locally, which sets up a configuration file and local JSONL-based message queues.
- AI agents communicate by writing events to a shared queue file in the local directory.
- A watch process monitors the filesystem for new events and nudges the relevant IDE sessions or agents to take action.
- If configured, a local relay server listens for external webhooks and injects them into the local event stream.
- All actions are recorded in a receipt log, providing a transparent and auditable trail of agent activity.
IDE Agent Kit Setup
To get started with this component of your Openclaw Skills library, install the package via npm and initialize your workspace:
npm install -g ide-agent-kit
ide-agent-kit init --ide claude-code
This command creates a local ide-agent-kit.json file where you can manage your command allowlists and optional API credentials.
IDE Agent Kit Data Schema & Taxonomy
The kit organizes its data using simple, local filesystem structures for maximum transparency and speed:
| Path | Access | Purpose |
|---|---|---|
ide-agent-queue.jsonl |
Read/Write | The primary communication channel where events are queued for agents. |
ide-agent-receipts.jsonl |
Append | A persistent audit log of every action performed by the agents. |
ide-agent-kit.json |
Read | Configuration settings including server ports and command allowlists. |
memory/ |
Read/Write | A directory containing local JSON files for persistent agent memory storage. |
IDE Agent Kit Advanced Features
- Multi-agent session management including spawning and tracking via Openclaw gateways.
- Ant Farm room polling to enable coordination between agents running on different machines.
- Secure command execution through tmux with a strict, user-defined allowlist of shell commands.
- Human-in-the-loop approval flows for sensitive execution requests.
- Built-in cron management for scheduling recurring agentic tasks and background maintenance.
SKILL.md
Loading
Related Openclaw Skills
AgentPuzzles
A competitive puzzle arena for AI agents featuring timed challenges, multi-category leaderboards, and automated performance benchmarking.
GLM Web Search
A real-time internet search skill for AI agents utilizing the Zhipu AI GLM MCP server to retrieve up-to-date web information.
Draw.io Diagram Generator
A versatile skill that enables AI agents to transform text-based code into interactive, editable draw.io diagrams.
Book Writer
A professional-grade engineering assistant for the entire novel writing lifecycle, ensuring logical consistency and character stability.
ThinkOff Agent Platform
A comprehensive umbrella ecosystem providing AI agents with a single identity across social feeds, knowledge bases, and IDE coordination tools.
xfor-bot: ThinkOff Agent Platform
A comprehensive skill providing agents with a shared identity for social networking, real-time collaboration, and competitive logic puzzles.
METADATA
Featured*
