菜单
Default workflow for smmariquit repositories. Use at the start of any task — read README, match conventions, minimal diffs, no secrets in commits, verify before redoing infra work. Check for stack-specific skills in this repo (.cursor/skills/).
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Anti-AI-slop design skill for greenfield pages, audits, redesigns, and design extraction from URLs or screenshots. Use when the user asks to build a new app or landing page, wants to redesign something, invokes Hallmark by name, or uses audit/redesign/study. Room TBA: also read references/room-tba.md before map chrome, sidebar, or editor UI work.
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.
Write structured handoff prompts when ending a session or delegating to another agent (Claude, Codex, Cursor). Use when user says handoff, delegate, continue in new chat, or epic spans multiple sessions.
Review diffs for over-engineering only — delete, stdlib, native, yagni, shrink. Use after a fat agent diff, before merge, or when user says ponytail-review, simplify review, what can we delete. Codex/Claude: use @ponytail-review plugin skill when installed.
Louis Rossmann anti-AI-slop rules for prose (README, docs, MDX, PRs, blog). From realrossmanngroup/no_ai_slop_writing_rules. Use when writing or editing any user-facing text. Not for code generation.
Implements and debugs Room TBA contributor edit proposals (suggest an edit, review queue, approve/reject, create proposals). Use when working on edit_proposals, proposal-service, Suggest an edit UI, admin review routes, proposal edge cases, or issue
| name | repo-baseline |
| description | Default workflow for smmariquit repositories. Use at the start of any task — read README, match conventions, minimal diffs, no secrets in commits, verify before redoing infra work. Check for stack-specific skills in this repo (.cursor/skills/). |
README.md (and CONTRIBUTING.md if present)..cursor/skills/ — load any skill that matches the task.package.json / pyproject.toml / Cargo.toml for scripts before inventing commands.--no-verify unless the user explicitly requests it.If the task touches Cloudflare, Vercel, or DNS across multiple repos, check smmariquit/tutorials for MIGRATION-HANDOFF.md and MIGRATION-STATUS.md before redoing migration work.
For Cloudflare or Vercel tasks: try the enabled MCP tools first (after auth). Fall back to gh, wrangler, or curl if MCP lacks scope or returns auth errors — report which path was used.