JasonColapietro/suede-creator-skills

README preview

# Suede Creator Skills

A 67-skill toolkit for Claude Code and Codex: take broad outcomes full send, orchestrate multi-agent teams and OpenAI Codex CLI worker fleets, run code review with an A-F ship grade, and design AI evals.

![License: MIT](https://img.shields.io/badge/License-MIT-blue) ![Skills: 67](https://img.shields.io/badge/Skills-67-black) [![GitHub stars](https://img.shields.io/github/stars/JasonColapietro/suede-creator-skills?style=social)](https://github.com/JasonColapietro/suede-creator-skills/stargazers)

> **By [Jason Colapietro](https://suedeai.ai/founder) / [Suede Labs AI](https://suedeai.ai)**

## What it is

A free, MIT-licensed, broadly reusable pack of **67 public skill folders** for Claude Code and OpenAI Codex. Each skill is a `skills/<name>/SKILL.md` file the agent loads on demand.

- **Full Send**: turn max-effort, max-agent, spare-no-compute, and fix-everything intent into one authorized controller, every useful non-colliding lane, adversarial reconciliation, and concise proof (`suede-full-send`). House line: **"Never end your allocation above zero."** Dry joke, not a literal token promise.
- **Agent orchestration**: wire complex changes into coordinated agent lanes with WIP collision detection, RFC mode, feature-flag strategy, rollback trees, and a handoff checklist that won't close without evidence (`suede-agent-teams`).
- **Codex worker fleets**: the Suede Fable Fleet — a Claude orchestrator decomposes a high-volume job, writes self-contained briefs, spawns parallel OpenAI Codex CLI `codex exec` workers, and reviews every output against acceptance criteria before anything ships (`suede-codex-fleet`).
- **Code review + A-F ship grade**: deep findings plus a blunt letter verdict across 7 evidence-backed lanes, with instant-F triggers and grade caps for auth and payment surfaces (`suede-code`, `suede-code-review`, `suede-code-grader`, `suede-ship-gate`).
- **AI evaluation**: turn LLM, RAG, classifier, and agent surfaces into AI-SPEC artifacts, failure-mode rubrics, eval cases, and acceptance gates (`suede-ai-eval`).
- **Next-action recommendation**: score 2-4 candidate moves on goal alignment, unblocking, evidence, urgency, and leverage, then return one recommendation packaged as a short runnable prompt (`suede-recommend-next-action`).
- **Design, copy, and SEO**: design systems and visual QA, conversion copy with an anti-slop gate, SEO/AEO/AI-EO audits, and A-F page visibility grades (`johnny-suede-design`, `johnny-suede-write`, `suede-design`, `suede-copy`, `suede-seo-audit`, `suede-visibility-grader`, `suede-site-alchemy`).
- **Anti-slop pass**: strip AI writing tells from finished prose before it ships: filler phrases, false agency, formulaic contrasts, passive voice, and metronomic rhythm, all against a merged kill list with a 50-point score gate (`suede-deslop`).

View full repository on GitHub →