A technical toolset for extracting architectural patterns, tool definitions, and system prompts from GitHub repositories and npm packages.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install source-analysis
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 source-analysis 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 Source Analysis?
Source Analysis is a specialized capability designed for developers to perform deep technical audits of AI-driven projects and agentic systems. By leveraging automated collection and binary extraction techniques, this tool uncovers the underlying logic, prompt engineering strategies, and permission models used by industry-leading agents. It serves as a vital component for those building their library of Openclaw Skills to benchmark or research modern software architecture and agentic design patterns.
Source Analysis Use Cases
- Benchmarking the architecture of AI coding assistants like Claude Code, Cursor, or Copilot.
- Reverse engineering system prompts and behavioral instructions from compiled binaries or distributed packages.
- Analyzing tool definitions and input/output schemas within Agentic SDKs to understand integration patterns.
- Researching memory management, permission systems, and context handling in open-source projects.
How Source Analysis Works
- Information Gathering: The skill explores GitHub repositories by fetching READMEs and recursive file trees using raw content URLs and APIs.
- Package Acquisition: It downloads and unpacks npm or PyPI packages to inspect distribution structures and platform-specific binaries.
- Binary Extraction: It uses advanced grep patterns to identify identity declarations, tool names, and system prompt segments embedded within executables.
- Schema Analysis: The tool parses TypeScript definition files to map out tool names, permission models, and concurrency safety flags.
- Report Generation: All extracted data is synthesized into a structured Markdown analysis report for architectural review.
Source Analysis Setup
To integrate this capability into your Openclaw Skills workflow, ensure your environment has access to standard CLI utilities. Basic commands used by the skill include:
# Download and unpack an npm package for analysis
npm pack @scope/package-name
tar xzf package-name-*.tgz
# Extract identity declarations from a binary file
grep -aoP 'You are [^"]{0,500}' /path/to/binary | sort -u
Source Analysis Data Schema & Taxonomy
The skill organizes findings into structured reports and temporary workspaces:
| Component | Details |
|---|---|
| Workspace Report | Saved as {workspace}/source-analysis-{project-name}.md in structured Markdown. |
| Temporary Data | Package fragments and binary stubs are stored in /tmp/ for processing. |
| Report Sections | Includes Architecture Overview, System Prompts, Tool Definitions, and Design Patterns. |
| Schema Formats | Extracted tool schemas are represented in JSON or Markdown tables for clarity. |
Source Analysis Advanced Features
- Heuristic-based detection of feature flags and environment configuration parameters.
- Multi-stage binary inspection using targeted grep patterns to bypass missing strings commands.
- Automated mapping of agentic permission systems and execution classifiers.
- Correlation analysis between raw GitHub source and compiled distribution packages.
SKILL.md
Loading
Related Openclaw Skills
IFQ App Builder
A streamlined utility that transforms three-sentence natural language intents into verified, build-ready prompt bundles for mobile, desktop, and web applications.
Claim Verifier
An automated fact-checking tool that extracts factual claims from draft content and verifies them against real-world evidence using Prismfy.
Tomoviee Video
Tomoviee Video is an AI-driven cloud editing skill that transforms raw recordings into professional 1080p short movies through simple text commands.
Editor Opus
A hands-free AI video editor that transforms raw footage into professional clips with simple text commands.
MuHaven RWA Portfolio
A confidential real-world-asset portfolio agent designed to manage encrypted token primitives through a secure, multi-tiered confirmation framework.
find-her
An intelligent dating coach and relationship strategist designed to help users find compatible partners through realistic, respectful, and actionable planning.
Featured*
