Ghostclaw is an architectural code review and refactoring assistant designed to perceive system-level flow issues and maintain codebase integrity.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ghostclaw-clone
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 ghostclaw-clone 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 Ghostclaw?
Ghostclaw serves as an advanced architectural ghost for your development workflow, moving beyond simple linting to analyze the structural soul of your projects. As part of the Openclaw Skills ecosystem, it identifies energy flow within data paths, detects legacy cruft known as code ghosts, and suggests deep refactoring to improve cohesion and reduce coupling. It is designed to ensure that your codebase aligns perfectly with the philosophical best practices of your specific tech stack.
By sensing the weight of dependencies and module responsibility, Ghostclaw provides a unique perspective on maintainability. Whether used as a standalone CLI tool or a sub-agent, it helps developers transition from code that merely works to code that is architecturally sound and elegantly structured.
Ghostclaw Use Cases
- Providing architectural insights during complex code reviews where standard linters fall short.
- Identifying structural issues in modules that compile successfully but feel heavy or disorganized.
- Automating repository health monitoring to ensure long-term vibe health and architectural consistency.
- Generating automated pull requests for architectural improvements using GitHub integration.
- Benchmarking microservices to track architectural health trends across multiple repositories.
How Ghostclaw Works
- The tool initiates a scan of the target repository or directory to assess the overall vibe health and architectural integrity.
- It utilizes specialized engines like PySCN for dead code detection and AI-CodeIndex for AST coupling analysis.
- Ghostclaw evaluates the code against tech-stack specific heuristics (e.g., React component size or Go package cohesion).
- A comprehensive architectural diagnosis is generated, identifying structural flaws and heavy modules.
- The assistant provides a refactor blueprint and precise code-level suggestions, including patches for immediate improvement.
- For automated workflows, it can open GitHub PRs or generate timestamped Markdown reports for the development team.
Ghostclaw Setup
To integrate Ghostclaw into your Openclaw Skills workflow, follow these steps:
- Install the required Python dependencies:
npm run install-deps
- Create a
repos.txtfile containing the paths to the repositories you wish to monitor. - Set your
GH_TOKENenvironment variable to enable automated Pull Request creation. - Run an initial test scan on a repository:
python3 src/ghostclaw/cli/ghostclaw.py /path/to/repo
Ghostclaw Data Schema & Taxonomy
Ghostclaw organizes its architectural analysis and metadata using the following structure:
| File/Directory | Description |
|---|---|
ARCHITECTURE-REPORT-<ts>.md |
The primary output containing vibe scores and refactor plans. |
src/ghostclaw/core/ |
Core Python analysis engines. |
src/ghostclaw/stacks/ |
Logic for tech-stack specific rules and heuristics. |
stack-patterns.yaml |
Configurable architectural rules for different frameworks. |
compare.py |
Data for trend analysis across multiple repos. |
Ghostclaw Advanced Features
- Multi-agent support: Can be invoked as a sub-agent using the ghostclaw codename within the Openclaw Skills framework.
- Background Watcher: Support for cron-based monitoring to perform daily architectural health checks.
- Automated PR Engine: Automatically creates GitHub pull requests with detailed architectural improvement suggestions.
- Custom Engine Toggles: Explicitly enable or disable the PySCN or AI-CodeIndex engines via CLI flags.
- Tech Stack Adaptability: Dynamic heuristics for Node.js, React, Python, Go, and Rust.
SKILL.md
Loading
Related Openclaw Skills
geo-claw
geo-claw is a specialized AI agent designed to manage the full lifecycle of Generative Engine Optimization (GEO) to boost brand visibility in AI-powered search results.
Use Cases Page Generator
A specialized technical framework for generating high-converting, scenario-first use case pages that bridge the gap between product features and customer problems.
nft-monitor
An automated 24/7 monitoring tool designed to track rare NFT listings and significant price fluctuations across major platforms.
SLA Monitor
A production-grade monitoring toolkit designed to track uptime, response times, and incident response for AI agents and automated services.
Oil Price Monitor
An automated monitoring skill for tracking domestic refined oil price adjustments and notifying users to refuel before price hikes.
golang-lint
An expert Go code quality skill that automates linting, configuration, and issue resolution using golangci-lint.
Featured*
