Skip to content
Back to skills

Feat Setup

ASecurity

Use when the user wants the project instructions file generated or updated from the detected stack and conventions — 'gerar o arquivo de instruções do projeto', 'criar convenções do projeto', 'setup governance', 'gerar AGENTS'. Writes the shared instructions file read by Claude Code, Codex, Hermes and OpenCode. Do NOT use to analyze the codebase in depth (feat-map) or to initialize the workspace (feat-init).

  • 3 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 28, 2026
ai-agentspythongotestingsecurity

Works with

  • claude code

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 28, 2026

npx -y skills add pwdev-solucoes/pwdev-claude-marketplace --skill feat-setup --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Feat Setup?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Feat Setup
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/pwdev-solucoes-feat-setup/badge)](https://www.skillsdirectory.com/skills/pwdev-solucoes-feat-setup)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: feat-setup
description: >
  Use when the user wants the project instructions file generated or updated from the detected
  stack and conventions — 'gerar o arquivo de instruções do projeto', 'criar convenções do
  projeto', 'setup governance', 'gerar AGENTS'. Writes the shared instructions file read by
  Claude Code, Codex, Hermes and OpenCode. Do NOT use to analyze the codebase in depth
  (feat-map) or to initialize the workspace (feat-init).
metadata:
  version: 3.2.1
---

# Generate the project instructions file

Which files each runtime reads, and the write rules, are in `references/project-files.md` —
read it first. The canonical content comes from `templates/AGENTS.template.md`.

## Procedure

1. **Language** — resolve `lang` (`references/language.md`).
2. **Check existing** — for each instructions file named in `references/project-files.md`,
   report whether it exists and show its first 20 lines. Existing content → ask: update (merge
   sections), replace, or skip. Never overwrite silently.
3. **Gather context** — `.planning/feat/codebase.md` and `.planning/feat/context/*.md` when
   present (suggest `feat-map` first when absent), otherwise a quick read-only
   `python3 "<plugin-root>/scripts/feat_map.py" --repo-root .`.
4. **Ask (1 round max)** only what cannot be detected: what the project is (1 sentence) and any
   specific rules ("use defaults" is fine).
5. **Write** the canonical file from the template with real values — commands from
   `context/testing.md`, conventions from `context/conventions.md`. Then make Claude Code load
   the same rules as `references/project-files.md` prescribes (a one-line import file when it
   does not exist yet).
6. **Present** the sections written (Identity, Stack, Architecture, Conventions, Commands,
   Quality, Security, Golden Rules), the files created or updated, and next:
   `feat-feature "description"`.

## Prohibitions

- Never overwrite an existing instructions file without asking.
- Never invent conventions — detect them or ask.

Language: resolve `lang` per `references/language.md` before any human-facing output. Safety: never read or expose `.env*` (except `.env.example`/`.template`/`.sample`), keys, certificates or credentials — `references/safety.md`. Paths `references/`, `scripts/`, `templates/`, `schemas/` are relative to the plugin root (`references/runtime.md`).

Files in this skill

  • SKILL.md2.3 KB
  • agents/openai.yaml187 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…