What problem does it solve? New AI coding agents lack context about a codebase's build commands, architecture, and conventions, forcing developers to repeatedly explain project specifics in every session. ## Core Features & Use Cases - Codebase Exploration: Scans the repository for build/test commands, architecture decisions, conventions, and pitfalls to capture essential agent knowledge. - Instruction File Generation: Creates or updates chat customization files such as AGENTS.md or .github/copilot-instructions.md, preferring to merge with existing files rather than duplicating content. - Convention Discovery: Detects existing agent configuration files (CLAUDE.md, .cursorrules, .windsurfrules, and similar) and links to existing documentation instead of copying it. - Use Case: After cloning a new project, run this Skill to produce an AGENTS.md file that tells any AI coding agent how to build, test, and navigate the codebase. ## Quick Start Ask the agent to initialize chat customization files for this repository, optionally naming a focus area such as testing conventions.