project-summary

What This Skill Does

The project-summary skill serves as an automated technical auditor for your codebase. It acts as a bridge for new developers, AI agents, or maintainers who need an immediate, bird's-eye view of an unfamiliar repository. By performing a multi-step analysis—ranging from scanning dependency manifests (like package.json or pyproject.toml) to inferring architectural patterns based on directory structure—the skill synthesizes technical debt, project structure, and essential entry points into a clean, human-readable report.

Installation

To integrate this skill into your environment, use the OpenClaw CLI: clawhub install openclaw/skills/skills/fratua/project-summary

Use Cases

• Onboarding Developers: Instantly provide new team members with a roadmap of the project's architecture and key entry files without requiring a senior dev to explain the repo.
• Agent Context Injection: Use the output as a system prompt prefix when asking OpenClaw to perform complex refactors or feature implementations.
• Technical Due Diligence: Quickly audit a legacy project or an open-source repository to understand its stack, deployment mechanisms, and quality standards.
• Legacy Documentation: Generate initial documentation for projects that lack a README or where technical debt has obscured the intended project structure.

Example Prompts

1. "Analyze the current directory and provide a summary of our architecture and entry points so I can start working on the new authentication module."
2. "Run project-summary on this repo to help me identify where the business logic is separated from the infrastructure layer."
3. "Generate a project overview, focusing on the CI/CD pipelines and the core application entry points, to update our team's onboarding wiki."

Tips & Limitations

• Context Window: For extremely large, monolithic repositories, the skill relies on file tree analysis. Ensure your .gitignore is well-maintained, as the skill automatically excludes common ignored folders like node_modules or dist to avoid noise.
• Pattern Recognition: While highly accurate for standard Go (Go standard layout), Node.js (MVC/Next.js), and Python frameworks, custom or highly unconventional architectures may require manual oversight.
• Performance: On very large monorepos, the initial file scan may take a few seconds longer. Run it from the root directory for the most comprehensive results.