Generate CLAUDE.md and AGENTS.md documentation from codebase analysis.
npx skills add https://github.com/zackham/agent-skills --skill build-llm-docs
Please help me install this Agent Skill. Skill: build-llm-docs Source: https://github.com/zackham/agent-skills/tree/main/build-llm-docs Command: npx skills add https://github.com/zackham/agent-skills --skill build-llm-docs
This Skill automates the creation of comprehensive and actionable documentation (CLAUDE.md/AGENTS.md) for codebases, enabling AI agents to contribute effectively without extensive human onboarding.
Use the build-llm-docs skill to generate CLAUDE.md documentation for the current project.
High-intent search queries and answers about installing and using this skill.
To generate LLM documentation for a codebase, you can automate the creation of detailed instruction files like CLAUDE.md or AGENTS.md through a multi-phase process of scouting, deep dives, and consolidation. This derives specific runtime and testing instructions from codebase analysis.
An AGENTS.md file provides specific, executable instructions derived from codebase analysis to help AI agents understand and contribute to projects safely. You need it for developer onboarding when launching new features or integrating AI developers without extensive human intervention.
Yes, you can automate CLAUDE.md creation for large codebases using a multi-phase automated process. It employs scouting, deep dives, consolidation, and review stages to ensure thorough codebase analysis, capturing runtime, patterns, testing, data, and guardrails.
The best way to provide context to AI agents is by generating high-quality, actionable LLM documentation. This creates a mission briefing with specific instructions derived from codebase analysis, allowing the AI to modify the codebase safely and efficiently.
Yes, automated documentation generation works without dependencies. The process analyzes the codebase directly through its multi-phase scouting and deep dive stages to extract patterns, runtime requirements, and guardrails, requiring no external dependencies.
Limitations of generating LLM docs depend on the codebase's structure and clarity. While the multi-phase process covers runtime, patterns, testing, data, and guardrails, highly abstract or poorly structured code may result in less actionable instructions for AI agents.
darelabs-techMine brownfield software patterns via AST analysis to generate architectural documentation and JSON facts.
sameer-thakkarGenerate AGENTS.md, CLAUDE.md, and .claude/rules/ files from codebase analysis.
Ikki-GroupExtract reusable UI patterns, components, and design tokens from codebases.
georgekhananaevEnforce consistent Markdown syntax across documentation and README files.
