skills.rest
bm629
bm629COMMUNITY
Explore all skills by this author →
GitHub

reviewing-prd

Reviews finished PRDs against a 12-condition plannability bar and emits an approve or revise verdict.

1|1|Updated May 24, 2026
Search Tags:#requirements traceability#review#prd#product requirements#verdict#acceptance gate#plannability
One-click install
npx skills add https://github.com/bm629/agent-skills --skill reviewing-prd-bm629
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: reviewing-prd
Source: https://github.com/bm629/agent-skills/tree/main/skills/reviewing-prd
Command: npx skills add https://github.com/bm629/agent-skills --skill reviewing-prd-bm629

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Product teams need an independent acceptance gate that decides whether a finished PRD is complete enough to plan the downstream build from, without the author grading their own work or reviewers inventing inconsistent standards. ## Core Features & Use Cases - 12-Condition Plannability Checklist: Judges evidenced problems, concrete personas, measurable metrics with targets, defensible MVP boundaries, testable acceptance criteria, NFR targets, traceability, named dependencies, and more. - Fabricated Evidence Detection: Flags precise-but-unsourced statistics and untraceable citations, treating suspected fabrication as a blocking revise. - Delta-Scoped Amend Review: Reviews only the change and its ripple when judging an amended PRD, rather than re-reviewing untouched sections. - Machine-Parseable Verdict: Emits exactly VERDICT: approve or VERDICT: revise plus actionable findings naming the failed condition, location, and fix. - Use Case: After authoring a PRD for a new onboarding flow, run this review to confirm every feature traces to a goal and every metric has a target before handing the document to engineering for planning. ## Quick Start Use reviewing-prd on docs/specs/onboarding-prd.md and give me the verdict with any blocking findings.

Frequently Asked Questions about reviewing-prd

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review a PRD before engineering planning starts?▼

Run the review against the 12-condition plannability bar covering evidenced problems, measurable metrics, MVP boundaries, testable acceptance criteria, NFR targets, traceability, and dependencies. The output is a single VERDICT: approve or VERDICT: revise line plus actionable findings.

What makes a PRD ready for development planning?▼

A plannable PRD has an evidenced problem, concrete personas, metrics with targets and measurement methods, explicit non-goals and release criteria, testable acceptance criteria, numeric NFR targets, no orphan features, and named dependencies. Thin PRDs pass when every applicable condition is met.

Can this review an amended PRD instead of a new one?▼

Yes, amendments get a delta-scoped review under condition 12. It checks the change and its ripple effects, change history, and orphaned downstream references rather than re-reviewing the unchanged sections of the document.

Does it catch fabricated statistics in AI-written PRDs?▼

Yes, condition 6 treats precise-but-unsourced numbers and untraceable citations as suspected fabrication. Suspected invented evidence always triggers a revise verdict until the claim is sourced or restated as an explicit assumption.

When should I not use this PRD review skill?▼

Do not use it for authoring or fixing a PRD, reviewing engineering design documents like specs, ADRs, or RFCs, or checking template section conformance. It only judges a finished PRD's plannability against the fixed bar.

Related Skills

Explore More Product & Management Skills→
GeneralJerelGeneralJerel
6

create-prd

Draft structured PRDs from problem statements for product teams.

Community
Advanced
rodolfochiconerodolfochicone
19

rc-spec

Interviews users to produce PRD, tech spec, and ADR documents.

Community
Advanced
SantiagoisperSantiagoisper

ce-optimize

Run metric-driven iterative optimization loops with parallel experiments and LLM-as-judge scores.

Community
Advanced
YoungLeadersDotTechYoungLeadersDotTech
10

initiative-overview-template

Generate structured initiative overview documents with YAML frontmatter and defined sections.

Community
Basic