A comprehensive CLI integration for Openclaw Skills that enables autonomous code development, auditing, and repository management via Claude Code.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-claude-code-bailian
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 openclaw-claude-code-bailian 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 CLI Integration?
Claude Code is a high-performance command-line interface that allows developers to interact with their codebase using natural language. When integrated with Openclaw Skills, it provides a bridge for AI agents to perform complex engineering tasks such as multi-file refactoring, performance optimization, and automated pull request generation. This skill leverages the full power of Anthropic's models to act as a terminal-based pair programmer.
By utilizing Openclaw Skills, developers can choose between non-interactive execution for quick tasks or full PTY-backed interactive sessions for deep debugging. It supports various permission levels, ensuring that autonomous changes are only made when the developer grants explicit authority, making it a safe yet powerful tool for modern software engineering workflows.
Claude Code CLI Integration Use Cases
- Performing automated code reviews and analyzing git diffs for potential bugs.
- Refactoring legacy modules to use modern language features and performance patterns.
- Generating comprehensive unit tests and automatically fixing failing test cases.
- Streamlining DevOps workflows by creating descriptive git commits and PR descriptions.
- Mapping and understanding new or complex codebases through natural language queries.
- Integrating with external tools like GitHub, Jira, and Slack via the Model Context Protocol.
How Claude Code CLI Integration Works
- The agent initializes by verifying the authentication status of the Claude CLI within the Openclaw Skills environment.
- For simple queries, the agent uses the Print mode to execute a single command and capture the output.
- For complex tasks, the agent initiates an interactive PTY session, allowing for multi-turn reasoning and tool usage.
- The agent analyzes the local filesystem, identifies relevant files, and proposes a plan of action based on the permission mode.
- Changes are applied to the codebase, and the agent can optionally run shell commands to verify the fix or run linting.
- Finally, the agent can use git integrations to commit the changes and finalize the task.
Claude Code CLI Integration Setup
To get started with this skill for Openclaw Skills, install the Claude Code CLI using PowerShell or WinGet:
# Install via PowerShell
irm https://claude.ai/install.ps1 | iex
# Or install via WinGet
winget install Anthropic.ClaudeCode
Next, authenticate your session:
claude auth login --email [email protected]
Ensure your agent configuration has pty: true enabled to support interactive terminal features.
Claude Code CLI Integration Data Schema & Taxonomy
The skill manages its operation through local configuration and session metadata as detailed below:
| Component | Data Type | Purpose |
|---|---|---|
| Session History | Local Cache | Allows the agent to resume previous coding conversations using --continue. |
| Sub-agents | JSON Config | Defined in .claude/agents/ to specialize in tasks like security or documentation. |
| Settings | JSON | Located in %USERPROFILE%\\.claude\\settings.json for API endpoints and keys. |
| MCP Registry | Tool Schema | Connects the CLI to external services like GitHub or SQL databases. |
Claude Code CLI Integration Advanced Features
- Support for Model Context Protocol (MCP) to extend agent capabilities to third-party services.
- Specialized sub-agent delegation for focused tasks such as security audits or code reviews.
- Custom model provider support, allowing integration with platforms like Alibaba Cloud DashScope.
- Granular permission modes including 'plan' (read-only), 'medium' (confirmation required), and 'auto' (fully autonomous).
- Budgetary controls via
--max-budget-usdto prevent unexpected API costs during long-running tasks.
SKILL.md
Loading
Related Openclaw Skills
Discord Voice Memo Upgrades
A technical core patch for Clawdbot that resolves Text-to-Speech (TTS) failures by optimizing payload delivery for Discord voice memos.
OpenClaw AntSeed P2P Integration
A skill that enables OpenClaw to access decentralized AI models by routing requests through the AntSeed peer-to-peer network.
Tribunal: Quality Enforcement for AI Agents
Tribunal is a robust quality enforcement plugin for Claude Code that automates TDD, secret scanning, and multi-agent quality gates.
Tribunal Install & Initialise
A powerful quality enforcement plugin for Claude Code that automates TDD, secret scanning, and context monitoring for AI-driven projects.
API Credentials Hygiene
A technical auditing skill designed to harden API credential handling through environment variables, rotation plans, and least-privilege access.
CPC/MPQC Training & Competence Tracker (UK)
An automated tracking system for UK driver CPC and MPQC compliance that generates audit-ready matrices, reminders, and reports.
Featured*
