A powerful AI agent skill to automatically research codebases and generate structured architectural documentation with Mermaid diagrams.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install describe-design
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 describe-design 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 Describe Design?
The Describe Design skill is a sophisticated tool designed for Openclaw Skills that enables AI agents to perform deep architectural research on existing codebases. It bridges the gap between raw source code and high-level understanding by synthesizing complex logic into structured markdown documentation. This skill is essential for maintaining a living record of system design, ensuring that both human developers and other AI agents can quickly grasp how specific features or components function without wading through thousands of lines of code.
By leveraging this capability within the Openclaw Skills ecosystem, teams can significantly reduce the overhead associated with technical debt and knowledge silos. Instead of manually mapping out dependencies, the agent follows a rigorous multi-stage workflow to identify entry points, trace data flows, and document configuration options, resulting in a source of truth that stays relevant even as the implementation evolves.
Describe Design Use Cases
- Onboarding new developers by generating comprehensive feature walk-throughs and structural overviews.
- Documenting complex legacy systems or specific components before starting a major refactoring process.
- Creating standardized architectural documentation including data flows and component interactions for compliance or audits.
- Generating AI-readable system summaries to improve the context-window efficiency of other Openclaw Skills.
- Producing Mermaid diagrams to visualize request lifecycles and service dependencies for stakeholders.
How Describe Design Works
- Scope Definition: The agent asks the user to identify the specific feature, system, or component that requires documentation and clarifies the target audience.
- Initial Exploration: The agent performs a broad scan of the directory structure, READMEs, and entry points to build a mental model of the codebase.
- Scope Confirmation: A high-level outline is presented to the user for approval to ensure the deep research phase remains targeted.
- Deep Research: The agent traces code paths, identifies component interactions, locates data persistence layers, and builds a code reference index.
- Document Drafting: Using a standardized template, the agent generates a markdown file featuring Mermaid diagrams, component lists, and data flow descriptions.
- Finalization: After user review and location confirmation, the agent writes the finalized architectural document to the repository.
Describe Design Setup
To activate this skill within your Openclaw Skills configuration, ensure your AI agent has read access to the target repository. No additional third-party dependencies are required, though a markdown viewer with Mermaid support is recommended for viewing diagrams.
# Trigger the skill by asking your agent to document a system
"Research the payment processing module and create a design document."
The agent will then guide you through the stage-based workflow to capture the necessary architectural details.
Describe Design Data Schema & Taxonomy
The skill organizes its findings into a highly structured Markdown format optimized for both humans and machines. It avoids volatile data like line numbers to ensure longevity.
| Section | Contents | Format |
|---|---|---|
| Overview | High-level summary of the system purpose. | Markdown Paragraphs |
| Architecture | Visual representation of component relationships. | Mermaid Flowchart/Sequence |
| Components | Purpose, file location, and key functions per module. | Markdown Headers |
| Data Flow | Description of input-to-output transitions. | Narrative List |
| Configuration | Environment variables and configuration keys. | Code References |
| Code References | Table mapping components to file paths and symbols. | Markdown Table |
| Glossary | Definitions of project-specific terms. | Markdown Table |
Describe Design Advanced Features
- Intelligent "Happy Path" Tracing: Follows a typical request or action from entry point to completion without logic gaps.
- Stable Reference System: Uses relative repository paths and function symbols instead of line numbers to survive code refactors.
- Automated Mermaid Visualizations: Generates dynamic flowcharts and sequence diagrams to represent complex logic visually.
- Multi-Agent Compatibility: Produces documentation structured specifically to serve as context for other Openclaw Skills.
- Interactive Refinement: Includes built-in checkpoints for users to adjust the research scope before the agent performs deep file analysis.
SKILL.md
Loading
Related Openclaw Skills
Critical Code Reviewer
An adversarial AI code reviewer designed to ruthlessly identify security flaws, technical debt, and bad practices across multiple programming languages.
MyVibe Publish
A comprehensive automation tool for publishing static web content, directories, and ZIP archives directly to the MyVibe platform.
Web Scraper Pro (zhuyu28)
An advanced browser automation skill for complex data extraction, form filling, and UI interaction.
NeuralMemory
A biologically-inspired associative memory system that gives AI agents persistent, intelligent recall using spreading activation graphs.
LinkedIn Automator
An automated LinkedIn growth tool for content creation, scheduling, and performance tracking using browser-based workflows.
Crypto Research Framework
A comprehensive AI-driven investment research framework for generating professional crypto market reports and real-time monitoring.
Featured*
