A hierarchical documentation system that allows AI agents to navigate and understand complex codebases without reading every source file directly.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentlens
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 agentlens 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 AgentLens?
AgentLens provides a structured navigation layer for large-scale software projects, enabling users to grasp project architecture quickly through a multi-level indexing system. By generating specialized documentation for modules, symbols, and dependencies, it empowers developers and AI agents to work more effectively within the Openclaw Skills ecosystem. This approach reduces the need to process massive raw source files by providing pre-indexed maps and outlines that highlight the most critical parts of the code.
The tool acts as a lens that focuses on different layers of a project, from the high-level module organization down to the granular business rules and TODOs. Using Openclaw Skills like AgentLens ensures that your AI coding assistants have the context they need to make informed decisions without getting lost in the noise of a large repository.
AgentLens Use Cases
- Exploring a new or unfamiliar codebase to understand its architectural overview.
- Locating specific symbols, functions, or classes within massive files without full-text scanning.
- Identifying project-wide TODOs, warnings, and business rules before implementing changes.
- Mapping file dependencies and module relationships to assess the impact of a refactor.
How AgentLens Works
- Scan: The tool analyzes the project directory to generate a high-level INDEX.md as the entry point.
- Module Mapping: It breaks the project into logical modules, creating dedicated MODULE.md files for detailed views.
- Symbol Extraction: Hierarchical outlines are created for large files to facilitate fast symbol location.
- Knowledge Persistence: TODOs, warnings, and business logic are extracted into specialized memory.md files.
- Targeted Access: The AI agent or developer uses these generated documents as a map to locate and read only the necessary source code sections.
AgentLens Setup
To integrate AgentLens into your development workflow with Openclaw Skills, run the following command in your project root:
# Generate or refresh the documentation hierarchy
agentlens
This will create a .agentlens/ directory containing the hierarchical documentation. Always ensure this directory is up to date before starting an AI-assisted coding session.
AgentLens Data Schema & Taxonomy
AgentLens organizes data into a tiered hierarchy within the .agentlens/ folder to facilitate efficient lookup via Openclaw Skills:
| Level | Component | Description |
|---|---|---|
| L0 | INDEX.md |
The master project map and list of all modules. |
| L1 | modules/{slug}/MODULE.md |
Detailed overview of a specific module and its files. |
| L1 | modules/{slug}/outline.md |
Map of symbols, functions, and classes for large files. |
| L1 | modules/{slug}/memory.md |
Centralized list of TODOs, warnings, and business rules. |
| L1 | modules/{slug}/imports.md |
File-level dependency mapping within the module. |
| L2 | files/{slug}.md |
Deep-dive documentation for particularly complex files. |
AgentLens Advanced Features
- Hierarchical navigation flow from project-level overview down to specific source lines.
- Specialized memory tracking for persistent business rules and developer warnings.
- Automated symbol outlining for large files to reduce token consumption during AI analysis.
- Dependency mapping via import analysis to visualize codebase connectivity.
- Compatibility with Openclaw Skills to streamline automated code understanding across various AI agents.
SKILL.md
Loading
Related Openclaw Skills
Video Subtitles
A powerful tool for generating SRT subtitles, translating audio, and burning hardcoded captions into videos using specialized Hebrew and English AI models.
Personality Dynamics
A dynamic evolution system that allows OpenClaw agents to learn interaction patterns, adapt their tone, and switch between specialized personas.
llms-txt-generator
An automated crawler and conversational agent that generates properly formatted llms.txt files to make websites readable and transactable for AI agents.
Transition Multisport Coach
An AI-powered endurance coaching skill providing personalized training plans and performance analytics for multisport athletes.
Venice AI Media
A comprehensive toolset for high-fidelity, uncensored AI image and video generation powered by the Venice AI platform.
Generate Presentation Skill
An automated workflow for generating designer-grade presentations from simple markdown, web URLs, or topics.
Featured*
