本文へ移動
cccskills
無料GitHub で公開

spec-reviewer

Quality gate for product specifications before stakeholder review. Use this skill when: - A spec draft is ready and you want to validate it before sending to Engineering, Design, and QA - You want to check a spec against your template's completeness rules - You need to catch vague language, missing sections, or gaps before review - You want a pass/fail scorecard with specific fix suggestions

インストール方法を見る

含まれるファイル(1)

  • SKILL.md3.5 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Spec Reviewer

Validate a product specification against quality standards. Outputs a pass/fail scorecard with actionable fix suggestions — designed to run before sending specs for cross-functional review.

Workflow

  1. Read the spec — Read the target spec file
  2. Read the template — Cross-reference against your spec template
  3. Run all checks — Apply every check category below
  4. Generate scorecard — Output results table with pass/fail per check
  5. List fixes — For each FAIL, provide a specific, actionable fix suggestion

Check Categories

1. Section Completeness

Verify every required section exists and is not empty or placeholder-only. Common required sections:

  • Objective/Problem Statement
  • Scope (in-scope AND out-of-scope)
  • User Personas
  • Functional Requirements
  • User Stories (all 4 paths: happy, edge, error, empty)
  • Non-Functional Requirements
  • Success Metrics

2. Vague Language Detection

Flag these patterns (each is a WARN):

Vague adjectives/adverbs:

  • "fast", "quick", "slow" -> specify milliseconds
  • "user-friendly", "intuitive" -> specify interaction steps
  • "soon", "later" -> specify date or phase
  • "large", "small", "many" -> specify quantities
  • "etc.", "and so on" -> enumerate all items

Vague verbs:

  • "handle", "process", "manage" -> specify the exact operation
  • "support" -> specify what exactly and how
  • "improve", "enhance" -> specify the metric and target

3. Requirement IDs

  • Every user story has a unique ID
  • Every acceptance criterion has a unique ID
  • No duplicate IDs
  • Sequential numbering

4. Story Coverage

Each functionality block should have stories covering:

  • Happy path
  • Edge cases
  • Error states
  • Empty states

5. Consistency Checks

  • Same concept uses same term throughout
  • Referenced sections actually exist
  • Date format is consistent

Scorecard Output Format

# Spec Review Scorecard

**Spec:** {file path}
**Reviewed:** {date}
**Overall:** {PASS / FAIL / PASS WITH WARNINGS}

## Results

| # | Check | Section | Status | Details |
|---|-------|---------|--------|---------|
| 1 | Completeness | Objective | PASS | Answers who, what, why |
| 2 | Completeness | Scope | FAIL | Out-of-scope table empty |
| 3 | Vague language | Requirements | WARN | "fast response" needs ms target |

## Summary
- **Total checks:** {N}
- **PASS:** {N}
- **FAIL:** {N}
- **WARN:** {N}

## Fix List (Ordered by Severity)

### FAIL — Must Fix
1. **{Section} — {Issue}**
   {Specific fix instruction}

### WARN — Should Fix
1. **{Section} — {Issue}**
   {Specific fix instruction}

Grading Rules

  • PASS: Zero FAILs, zero or more WARNs
  • PASS WITH WARNINGS: Zero FAILs, 1+ WARNs
  • FAIL: 1+ FAILs

Anti-Patterns

  • Don't pass specs with empty out-of-scope sections — scope creep follows
  • Don't accept "TBD" in critical sections — flag as FAIL
  • Don't ignore vague language — it causes misalignment during build

Quality Checklist

  • All required sections checked
  • Vague language scan completed
  • Story coverage verified (4 paths)
  • Requirement IDs validated
  • Fix list is specific and actionable

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Rate any product, feature, or experience on the 11-star scale (Brian Chesky's Airbnb thought experiment). Use when user says "rate this experience", "11-star", "star rating", "experience audit", "how good is this", "experience rating", "product audit", "quality assessment", or wants to evaluate product quality and identify improvement paths. Also trigger when user wants to benchmark a feature, assess where a product stands, or map out what "great" looks like - even if they don't explicitly say "11-star".

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Generate A/B copy variants grounded in Eugene Schwartz's awareness and market-sophistication stages and John Caples' tested-headline discipline. Every variant declares its angle, the awareness stage it targets, and a falsifiable hypothesis, and the skill recommends which pair to test first and what delta counts as signal. Use when the user says "A/B test this", "write variants", "give me options", "test different angles", "copy variants", "ab copy", "test this headline", "multiple versions", "which version is better", or wants to test copy before publishing. For reviewing copy that already exists, see copy-review. For sample size and stopping rules, see growth-experiment. For diagnosing a page rather than its words, see page-cro. For full paid ad sets by platform, see ad-campaign-writer.

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Write measurable acceptance criteria for requirements or user stories. Use when the user says "write acceptance criteria", "definition of done", "how do we test this", "fit criteria", "given when then", "Gherkin scenarios", "what does done look like", "testable conditions", "how will we know this works" - even if they don't explicitly say "acceptance criteria".

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Build a strategic account plan for winning, retaining, or expanding a named account. Use when the user says "build an account plan", "strategic account planning", "how do I grow this account", "account expansion strategy", "whitespace analysis", "QBR prep", "key account review", "map the stakeholders in this account", "land and expand plan", "how do I break into [company]", or wants a structured strategy for a specific account.

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Write paid ad copy for LinkedIn, Google, Meta, and YouTube. Produces multiple variants per platform, each targeting a different Eugene Schwartz awareness level and Cialdini persuasion principle. Use when the user asks for ad copy, ad variants, "write LinkedIn ads", Google ads, Meta ads, paid social copy, ad headlines, or wants creative for a paid campaign. Reads brand voice, ICP, and positioning from knowledge/. For organic social posts, see linkedin-post or social-calendar. For the landing page the ad points at, see landing-page-writer. For testing variants, see ab-copy-writer.

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Design clean and consistent APIs. Use when the user says "design an API", "API design review", "design these endpoints", "REST API for X", "GraphQL schema", "API contract", "how should this endpoint work", "review my API design", or wants to create or improve an API interface - even if they don't explicitly say "API design".

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

qa-aman のスキルをすべて見る

このスキルの問題を報告する