create-styleguide
Per-surface visual styleguide lifecycle — capture, annotate with token callouts, audit via /review-design, track token debt, and detect visual drift between rounds. Composes /create-design-board and /review-design into a repeatable pipeline.
Informations de source
- Dépôt
- grahama1970/agent-stack-public
- Dernière activité de la source
- 24 septembre 2026 à 15:51
- Langue détectée de SKILL.md
- anglais
- Étoiles
- 0
- Forks
- 0
Options d'installation
Le prompt qui vérifie d'abord la source est sélectionné par défaut. Vous pouvez passer à une commande directe ou télécharger une copie locale.
Vérifiez les fichiers source
Lisez SKILL.md et les fichiers associés affichés par SkillsMP avant de décider de l'installer.
Explorateur de fichiers
9 fichiersAffichage de SKILL.md
SKILL.md
Instructions source · Aperçu en lecture seule- name
- create-styleguide
- version
- 1.0.0
- description
- Per-surface visual styleguide lifecycle — capture, annotate with token callouts, audit via /review-design, track token debt, and detect visual drift between rounds. Composes /create-design-board and /review-design into a repeatable pipeline.
- provides
- ["styleguide","visual-audit","token-debt-tracking"]
- composes
- ["create-design-board","review-design","memory","agentic-evals"]
- triggers
- ["create styleguide","build styleguide","visual styleguide"]
- allowed-tools
- ["Bash","Read","Write"]
- taxonomy
- ["design","compliance","visual"]
- disciplines
- ["ui-design-engineering"]
> STOP. READ THIS ENTIRE SKILL.MD BEFORE CALLING ANY ENDPOINT.
# /create-styleguide
Per-surface visual styleguide lifecycle for Embry OS's 10 UX surfaces.
## Commands
- `build` — Full pipeline: annotate -> audit -> debt -> assemble STYLEGUIDE.md
- `annotate` — Just the Pillow annotation step (token callouts on screenshots)
- `audit` — Run /review-design on a surface
- `diff` — Visual regression between rounds (pixel diff + block-SSIM)
- `status` — Show which surfaces have styleguides, debt counts, last audit dates
- `dry-run` — Generate sample output without LLM calls (used by sanity.sh)
## Usage
```bash
./run.sh build --surface S4 --skip-audit
./run.sh annotate --screenshots ./shots/ --tokens ./tokens.json --output ./annotated/
./run.sh diff --surface S4 --before ./round1/ --after ./round2/
./run.sh status
```
Voir sur GitHub