tmolavi

Agent Project Discovery Skill — AI skill for Claude Code

AI community

Universal startup and context discovery skill for AI coding agents (Claude Code, Antigravity, Cursor, Codex, Copilot).

How to install Agent Project Discovery Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open tmolavi/agent-project-discovery-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agent Project Discovery Skill does

Universal startup and context discovery skill for AI coding agents (Claude Code, Antigravity, Cursor, Codex, Copilot).

Alternatives in AI

Hexstrike AI — HexStrike AI MCP Agents is an advanced MCP server that lets AI agents (Claude, GPT, Copilot, etc.) autonomousl 8.2k ★
Codesight — Universal AI context generator 979 ★
Engrim — The Universal Cross-Model Episodic Memory Standard 219 ★

README

agent-project-discovery-skill 🧭

[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT) [![Release](https://img.shields.io/badge/release-v1.1.0-emerald.svg)](https://github.com/tmolavi/agent-project-discovery-skill/releases) [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/tmolavi/agent-project-discovery-skill/pulls) [![Supported Agents](https://img.shields.io/badge/Agents-Codex%20%7C%20Claude%20Code%20%7C%20Cursor%20%7C%20Antigravity%20%7C%20Copilot-blueviolet.svg)](#supported-ai-agents)

**"An AI agent should understand before it changes."**

🌍 Translations

[English](README.md) | [فارسی](README.fa.md) | [Türkçe](README.tr.md) | [Azərbaycanca](README.az.md) | [العربية](README.ar.md)

A universal, production-grade startup skill for AI coding agents. It prevents AI agents from jumping blindly into edits, clobbering existing architectures, asking redundant questions, or rewriting working code by teaching them to act like seasoned senior engineers onboarding onto an established codebase.

🎯 Why This Skill Exists

Today's AI coding assistants are capable, but their default initialization behavior often suffers from critical failure modes:

❌ Default Agent Pitfall	✅ With `agent-project-discovery-skill`
Eager Editing: Jumps straight into editing random files without knowing the tech stack or monorepo boundaries.	Structured Discovery: Explores repository structure, dependencies, configs, test setups, and documentation first.
Frivolous Questions: Interrupts developers asking "Where is X?" or "What should I do?"	Search First, Ask Later: Resolves questions autonomously via codebase grep, docs, and git history.
Clobbering Diffs: Overwrites uncommitted work or ignores ongoing work in the working tree.	Git Awareness: Inspects `git status`, current branch, recent commits, and protects existing user