Requirements Review Skill
Description
Problem — requirement reviews were inconsistent and depended on the reviewer's mood/experience Solution — formalized the criteria into a Claude Skill, tested and calibrated it Result — consistent reviews in minutes instead of an hour, a single standard across the team
Installation
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open the source below and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
README
requirements-review
A Claude Skill for reviewing requirements, PRs, and documentation for test-readiness.
When it triggers
Claude applies this skill when you ask to:
- review a requirement, PR, or piece of documentation
- do a "requirements review" / "PR review" / "documentation review"
- check whether a requirement, PR, or documentation is ready for testing
What it does
The skill reads the requirement and evaluates it against **7 criteria**:
| Criterion | What it checks |
|---|---|
| Atomicity | One requirement describes one behavior |
| Unambiguity | No vague terms ("fast," "convenient," "correct") without a definition |
| Completeness | Negative scenarios and edge cases are described |
| Consistency | No conflicts within the requirement or with other requirements |
| Non-duplication | No duplication of other requirements |
| Testability | A verification/test case can be written for every statement |
| Feasibility | The requirement can actually be implemented |
Every finding must be backed by an exact quote from the requirement text. If something is unclear or missing, it's raised as a **question for the BA**, not stated as a fact — the skill is instructed not to invent problems or business rules that aren't in the text.
Output format
Every review report has 4 sections, in this order:
- Verdict — ✅ Ready for test analysis / ⚠️ Needs clarification / ❌ Not ready
- Findings table — №, Criterion, Quote, Problem, Severity, Suggested fix
- Questions for the BA
- What's good — 1–2 positive notes
Severity levels
- Critical — blocks test analysis (expected behavior can't be understood)
- Major — risk of missed or contradictory test scenarios
- Minor — inaccuracy that doesn't block work but could cause misreading
Guardrails (built into the skill)
To keep the review grounded and prevent false positives, the skill explicitly checks itself before finalizing a report:
- Don't flag a value range/condition as "incom
Related Skills
Auto Update
Pull the latest ECC repo changes and reinstall the current managed targets.
Ecc Guide
Navigate ECC's current agents, skills, commands, hooks, install profiles, and docs from the live repository su
Epic Claim
Claim an epic issue, stamp coordination state, and sync local ownership.
Epic Publish
Publish a validated epic update back to the issue and local cache.
Epic Review
Mark epic review requested, approved, or changes requested.
Epic Unblock
Sweep blocked epic issues and reopen anything whose dependencies are closed.