# Initialize

> >

- **Type:** Skill
- **Install:** `agentstack add skill-kilimcininkoroglu-cli-tweaks-initialize`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [KilimcininKorOglu](https://agentstack.voostack.com/s/kilimcininkoroglu)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [KilimcininKorOglu](https://github.com/KilimcininKorOglu)
- **Source:** https://github.com/KilimcininKorOglu/cli-tweaks/tree/main/wrongstack/skills/initialize

## Install

```sh
agentstack add skill-kilimcininkoroglu-cli-tweaks-initialize
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Initialize - Create AGENTS.md

## Action

Immediately scan the codebase using the files listed below and create or update AGENTS.md. Do not wait for further instructions. Start scanning now.

## Target File

AGENTS.md only. Do NOT create or modify CLAUDE.md.

## Required Header

```
# AGENTS.md
This file provides guidance to AI coding agents (WrongStack, Claude Code, Cursor, Codex, Gemini CLI, GitHub Copilot, Devin, Windsurf, Amp, Jules, Aider, VS Code, Zed, goose, RooCode, etc.) when working with code in this repository.
```

## Files to Scan

| Priority | Files                                                                | Purpose                    |
|----------|----------------------------------------------------------------------|----------------------------|
| High     | `README.md`, `PROJECT.md`, `CONTRIBUTING.md`                         | Project overview           |
| High     | `package.json`, `Makefile`, `Cargo.toml`, `go.mod`, `pyproject.toml` | Build commands             |
| High     | `CLAUDE.md`, `GEMINI.md`                                             | Existing AI rules to merge |
| High     | `.cursor/rules/`, `.cursorrules`                                     | Cursor AI rules            |
| High     | `.github/copilot-instructions.md`                                    | Copilot instructions       |
| High     | `.github/workflows/`                                                 | CI/CD build/test commands  |
| Medium   | `docker-compose.yml`, `Dockerfile`                                   | Container setup            |
| Medium   | `.env.example`, `config/`                                            | Configuration              |
| Medium   | `src/`, `lib/`, `app/` structure                                     | Architecture               |
| Medium   | `.gemini/settings.json`                                              | Gemini CLI config          |
| Low      | Linter/formatter configs                                             | Code style                 |
| Low      | `.aider.conf.yml`                                                    | Aider config               |

## What to Include

- Build/Run Commands (install, build, run, test)
- Single Test Execution (how to run individual tests)
- Architecture Overview (high-level structure)
- Key Patterns (codebase-specific conventions)
- Environment setup requirements (if present)
- Database/migration commands (if present)
- Monorepo workspace structure (if present)
- Security considerations (if present)
- PR/commit message guidelines (if present)
- Deployment steps (if present)

## What to Exclude

- Generic development practices
- Obvious instructions
- Complete file/directory listings
- Information easily found in config files
- Made-up sections like "Tips for Development" or "Support and Documentation"

## Behavior

- If AGENTS.md does not exist: create it directly
- If AGENTS.md already exists: analyze current content and update it directly
- If CLAUDE.md or other AI config files exist, incorporate their important parts
- If a monorepo is detected, suggest creating nested AGENTS.md files for each subproject
- You can ONLY edit or create AGENTS.md -- never touch CLAUDE.md

## Output Guidelines

- Keep under 500 lines
- Focused, actionable, scoped rules
- Write like clear internal documentation
- Use tables for command references
- No fabricated information
- Commands verified against project structure
- IMPORTANT: Always write in English only, regardless of conversation language

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [KilimcininKorOglu](https://github.com/KilimcininKorOglu)
- **Source:** [KilimcininKorOglu/cli-tweaks](https://github.com/KilimcininKorOglu/cli-tweaks)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** yes
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-kilimcininkoroglu-cli-tweaks-initialize
- Seller: https://agentstack.voostack.com/s/kilimcininkoroglu
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.