What problem does it solve? AI coding agents often lack context about a codebase's build commands, architecture, and conventions, causing them to make mistakes or waste effort. This Skill explores a repository and produces chat customization files that make agents immediately productive. ## Core Features & Use Cases - Codebase Exploration: Discovers existing conventions, build/test commands, architecture decisions, and project-specific pitfalls through parallel subagent exploration. - Instruction File Generation: Creates or updates AGENTS.md or .github/copilot-instructions.md following link-don't-embed and minimal-by-default principles. - Iterative Refinement: Merges with existing files, preserves valuable content, and suggests follow-up customizations like skills, hooks, and prompts. - Use Case: After cloning an unfamiliar repository, run this command to produce an AGENTS.md that tells any AI agent how to build, test, and navigate the project. ## Quick Start Ask the agent to initialize chat customization files for this repository, optionally naming a focus area such as the frontend or testing conventions.