Claude Code Controller for Openclaw

The Claude Code Controller is a robust integration for Openclaw Skills that enables seamless management of the Anthropic Claude Code CLI for autonomous software engineering.

rocherkong
v1.0.0
Mar 10, 2026
1
1.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install claudecode-controller

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install claudecode-controller 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 Claude Code Controller?

The Claude Code Controller is an advanced skill designed to provide full control over the Anthropic Claude Code terminal-based coding agent within the Openclaw environment. It allows developers to leverage high-reasoning AI models for complex programming tasks, including project-aware feature implementation and deep codebase analysis. By utilizing Openclaw Skills and the Agent Control Protocol (ACP), this tool creates a secure, isolated runtime for long-running coding sessions, ensuring that your primary agent interaction remains focused while heavy-duty engineering happens in the background.

Claude Code Controller Use Cases

  • Implementing full-stack features from natural language requirements.
  • Conducting automated code reviews and identifying performance bottlenecks.
  • Executing large-scale code refactoring for better maintainability.
  • Generating comprehensive documentation and unit tests for existing modules.
  • Debugging complex issues by analyzing project-wide context and logs.

How Claude Code Controller Works

  1. The user triggers a task using the /claudecode command within Openclaw Skills.
  2. The controller initializes an isolated ACP session to launch the Claude Code CLI.
  3. The agent scans the local project directory to build a structural and contextual map.
  4. Claude Code executes a multi-step engineering plan, utilizing tools for file reading, editing, and shell command execution.
  5. Upon completion, the controller syncs the changes and provides a summary of the results back to the main session.

Claude Code Controller Setup

To start using this skill within Openclaw Skills, first install the Claude Code CLI globally:

npm install -g @anthropic-ai/claude-code

Next, configure your API credentials:

export ANTHROPIC_API_KEY="your-api-key"

Finally, initialize your project by creating a .claude/settings.json file in your project root to define permissions and model preferences.

Claude Code Controller Data Schema & Taxonomy

The skill manages its operation through a specific file and directory structure:

Data Type Description Path
Configuration Model selection and tool permission settings. .claude/settings.json
Execution Logs Detailed logs of every CLI session and agent action. ~/.claude/logs/latest.log
Custom Tools User-defined scripts to extend agent capabilities. ~/.claude/tools/
Project Context Metadata and indexing info for the codebase. .claude/

Claude Code Controller Advanced Features

  • Multi-model support allowing users to switch between Claude 3.5 Sonnet, Opus, and Haiku based on task complexity.
  • Custom tool integration allowing the agent to execute proprietary bash scripts for specialized build or test workflows.
  • Project templates for Express, React, and Python to accelerate the setup of new microservices.
  • Isolated session management via ACP, preventing context contamination and allowing for concurrent engineering tasks.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*