메뉴
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when the user wants to edit, author, update, translate, fix, or bulk-modify the actual content stored in MDCMS — phrases like "edit my content", "do something with the content", "update a post", "change the body of X", "add a section to my page", "fix typos in this article", "rename a doc", "translate this page to French", "add a new blog post", "make a new page", "bulk-edit a field across all posts", "find a doc with frontmatter X", or "I want to write a draft". For changes to *documents themselves*. Not for schema (that's `mdcms-schema-refine`), not for sync mechanics or auth (that's `mdcms-content-sync-workflow`), not for components (that's `mdcms-mdx-components`).
Use when the user wants to add, edit, extend, or fix MDCMS content types, fields, or references — phrases like "add a new content type", "register a new type with MDCMS", "create a content model", "make me an author type", "add a tags field to posts", "add a new field to my CMS", "link blog posts to authors", "edit mdcms.config.ts", "the inferred schema is wrong", "the schema is missing X", "change the schema", or "author the content model". Triggers on any intent to evolve the content schema, including localizing a type or adding a `fieldTypes.reference()` between types.
Use when the user runs day-to-day MDCMS CLI operations — `mdcms pull`, `mdcms push`, `mdcms status`, `mdcms login`, `mdcms logout` — or asks "how do I keep my local markdown in sync with MDCMS", "how do I push changes", "how do I publish from the CLI", "rotate the API key", "what's the draft vs publish flow", "add MDCMS to CI / GitHub Actions". Also triggers on auth-failure symptoms — "I'm getting 401 / unauthorized / 'not logged in' from the CLI", "MDCMS_API_KEY isn't working", "credentials missing", "mdcms login first". Operational usage after init — not for first-time setup.
Use this skill when the user wants to fetch MDCMS content from their React/Next/Remix app, says things like "render MDCMS content in my app", "replace my filesystem markdown with MDCMS", "use @mdcms/sdk to load posts", "SSR MDCMS content", "fetch drafts in preview mode", or similar. Covers both replacing an existing filesystem-based content layer (brownfield) and writing fresh fetching code (greenfield), plus drafts vs published, SSR patterns, and revalidation.
Use this skill when the user wants to embed the MDCMS Studio UI inside their own React/Next/Remix app, says things like "add the Studio to my app", "mount Studio at /admin", "give my editors a visual editor in the host app", "embed MDCMS editor", or similar. Walks through installing `@mdcms/studio`, adding a catch-all route, wiring the prepared Studio config, setting `MDCMS_STUDIO_ALLOWED_ORIGINS` on the server, and verifying login.
Use when the user wants to register, wire up, or author a React/MDX component for MDCMS — phrases like "add a new component", "register this component with MDCMS", "register a component with the CMS", "add a Callout component", "register custom MDX components", "make my React component available in Studio", "wire a component to MDCMS", "my MDX uses custom components", "my MDX renders plain text for <Alert>", "<Callout> shows as a raw tag in Studio", or "the Studio preview doesn't render my component". Triggers on a generic "add a component" intent inside an MDCMS-aware repo (with `mdcms.config.ts`).
| name | requesting-code-review |
| description | Use when completing tasks, implementing major features, or before merging to verify work meets requirements |
Dispatch superpowers:code-reviewer subagent to catch issues before they cascade. The reviewer gets precisely crafted context for evaluation — never your session's history. This keeps the reviewer focused on the work product, not your thought process, and preserves your own context for continued work.
Core principle: Review early, review often.
Mandatory:
Optional but valuable:
1. Get git SHAs:
BASE_SHA=$(git rev-parse HEAD~1) # or origin/main
HEAD_SHA=$(git rev-parse HEAD)
2. Dispatch code-reviewer subagent:
Use Task tool with superpowers:code-reviewer type, fill template at code-reviewer.md
Placeholders:
{WHAT_WAS_IMPLEMENTED} - What you just built{PLAN_OR_REQUIREMENTS} - What it should do{BASE_SHA} - Starting commit{HEAD_SHA} - Ending commit{DESCRIPTION} - Brief summary3. Act on feedback:
[Just completed Task 2: Add verification function]
You: Let me request code review before proceeding.
BASE_SHA=$(git log --oneline | grep "Task 1" | head -1 | awk '{print $1}')
HEAD_SHA=$(git rev-parse HEAD)
[Dispatch superpowers:code-reviewer subagent]
WHAT_WAS_IMPLEMENTED: Verification and repair functions for conversation index
PLAN_OR_REQUIREMENTS: Task 2 from .ai/plans/deployment-plan.md
BASE_SHA: a7981ec
HEAD_SHA: 3df7661
DESCRIPTION: Added verifyIndex() and repairIndex() with 4 issue types
[Subagent returns]:
Strengths: Clean architecture, real tests
Issues:
Important: Missing progress indicators
Minor: Magic number (100) for reporting interval
Assessment: Ready to proceed
You: [Fix progress indicators]
[Continue to Task 3]
Subagent-Driven Development:
Executing Plans:
Ad-Hoc Development:
Never:
If reviewer wrong:
See template at: requesting-code-review/code-reviewer.md