Does this skill work with monorepos?

Yes, it includes specific placement rules for root directories and individual packages, ensuring each subdirectory has the appropriate level of detail without duplicating root information.

What happens to development insights learned during a session?

Learned insights and recurring issues are classified and organized into a gotchas.md file to ensure the AI and future developers don't repeat the same mistakes.

What is the primary difference between README.md and AGENTS.md in this skill?

README.md is the 'entry point for humans,' focusing on high-level overviews and installation, whereas AGENTS.md is the 'guide for AI agents,' containing deep technical context and project structure.

How does the skill maintain consistency between files?

The skill treats AGENTS.md as the Single Source of Truth. If it detects a mismatch in the project overview or commands, it auto-corrects the README.md based on the data in AGENTS.md.

Project Context Manager

Name: Project Context Manager
Author: masseater

bymasseater

0•

Learning & Documentation

Maintains documentation consistency by organizing project metadata into human-centric READMEs and AI-optimized AGENTS.md files.

This skill establishes a robust framework for project documentation by enforcing a Single Source of Truth (SSOT) via AGENTS.md. It distinguishes between documentation intended for human visitors (README.md) and technical context required for AI agents (AGENTS.md), ensuring that both stay synchronized. The skill automatically manages project structure, architecture details, and session-based insights while refactoring files like gotchas.md and CLAUDE.md to provide a seamless development experience for both developers and AI assistants like Claude Code.

Key Features

01Automates consistency checks and updates for command lists and project overviews

02Differentiates documentation targets for human visitors vs. AI development tools

030 GitHub stars

04Organizes session-learned insights into classified gotchas.md files

05Enforces AGENTS.md as the technical Single Source of Truth for AI assistants

06Standardizes monorepo documentation across root and subdirectory levels

Use Cases

01Standardizing documentation across large monorepos to ensure AI agents have clear boundaries

02Syncing technical architecture updates from developer notes to public-facing README files

03Capturing and classifying tribal knowledge and 'gotchas' during active coding sessions

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add masseater/claude-code-plugin should-be-project-context

For use in Claude.ai and ChatGPT

Download Skill