What is an AGENTS.md file?

An AGENTS.md file is a specialized markdown document placed at key directory boundaries that provides specific instructions, conventions, and context maps for AI agents to follow when working in that directory.

Is this documentation useful for human developers?

Absolutely. While optimized for AI consumption with concise bullets, these files serve as excellent 'at-a-glance' guides for humans to understand module ownership and local development workflows.

How does this skill help with monorepos?

It creates a root-level map of all modules and places technical guidance within each specific module folder, ensuring the AI understands boundaries and doesn't apply the wrong tech rules to the wrong project.

Does it replace my existing documentation?

No, it complements it. AGENTS.md files are meant to be high-signal and actionable; they should link to your more detailed documentation in the /docs folder for deep dives.

AGENTS.md Builder

Name: AGENTS.md Builder
Author: Masked-Kunsiquat

byMasked-Kunsiquat

0•

학습 및 문서화

Generates a structured hierarchy of AGENTS.md files to provide AI agents with scoped context, module maps, and cross-domain workflow guidance.

The AGENTS.md Builder skill automates the creation and maintenance of localized documentation files designed to guide AI agents through complex codebases. By establishing a hierarchy of AGENTS.md files—at the root for global rules and within specific modules for tech-specific instructions—this skill ensures that Claude has the precise context needed for any given task without cluttering its context window. It is particularly effective for monorepos, where it maps features to paths, defines cross-domain workflows like frontend-to-backend API calls, and establishes clear boundaries for MCP tool usage and safety constraints.

주요 기능

01Hierarchical documentation structure (root and nested AGENTS.md)

02Cross-domain workflow documentation (e.g., API contracts, auth flows)

03Monorepo module and feature mapping to file paths

040 GitHub stars

05Standardized templates for coding conventions and safety rules

06Scoped tech-specific guidance for backend, frontend, and infrastructure

사용 사례

01Standardizing repository conventions and workflow rules across distributed teams.

02Optimizing AI agent performance in complex monorepos with multiple service boundaries.

03Mapping high-level user features to specific code entrypoints for faster agent navigation.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add masked-kunsiquat/crm-orbit2 agents-md

For use in Claude.ai and ChatGPT

Download Skill