woos-prd-consistency-audit

name	woos-prd-consistency-audit
description	Dedicated analyze gate for PRD and UI brief consistency. Runs script extraction first, then semantic review with evidence-backed findings.
version	1.0.0
author	Hermes Profile
license	MIT
metadata	{"hermes":{"tags":["product","prd","audit","consistency","gate"],"related_skills":["woos-product-design-flow","woos-ui-design-brief"]}}

PRD Consistency Audit

Output Language

• Author all user-facing prose (narratives, findings, summaries) in the user's most recent input language (e.g. if the user is writing in Chinese, write the PRD body, review findings, and summaries in Chinese).
• Keep these tokens verbatim in English regardless of the user's language, because they are contract identifiers consumed by other skills and gates:
  • Verdict tokens: PASS, REQUEST_CHANGES, BLOCKED, NOT_RUN
  • Severity tokens: critical, high, medium, low, warning
  • Tag tokens: [ASSUMPTION: ...], [NEEDS CLARIFICATION: ...], [NEEDS INPUT: ...]
  • Phase / dimension IDs: P0 … P11, P2a, Phase A, Phase B
  • Shape values: internal-tool, single-operator, consumer-product, multi-stakeholder, CLI
  • P0 rating values: strong, adequate, thin, broken
  • Section headings declared in templates (e.g. ## Background, ## Functional Requirements, ## Assumptions Index) — translations break downstream structural checks
  • Field labels inside templates (e.g. **Consequences (testable):**, **Out of Scope:**, **User value:**, Given … When … Then …)
• Code, file paths, IDs, and quoted PRD excerpts stay in their original form.

Purpose

Run Step 6 of woos-product-design-flow as an isolated audit skill.

This skill exists to stop the parent orchestrator from skipping straight to a soft summary like "looks consistent enough."

Required Invocation (hard gate)

• MUST be invoked as a separate skill
• MUST run scripts/analyze_gate.py before semantic review
• MUST use the script output as evidence, not as the final verdict
• MUST review every A1-A5 check explicitly
• If script execution is skipped or any check is omitted, return BLOCKED

Required Load Set (mandatory)

Before auditing, load and report:

• references/framework-implementation-readiness.md
• scripts/analyze_gate.py
• docs/prd/<version>/<feature-id>.md
• docs/design/<version>/<feature-id>-ui-brief.md (if present)

If any required file is not loaded, return BLOCKED.

Conditional Load Set (upstream dependencies)

When the orchestrator provides upstream interface summaries, also load:

• docs/prd/<version>/<upstream-feature-id>-interface.md for each declared upstream dependency

When upstream interface summaries are present, add A6 to the required checks:

| A6 | Upstream Interface Consistency | All shared concepts used in this feature's PRD/UI match upstream interface definitions (enums, field names, event types, terminology) |

A6 failures count toward GAPS_FOUND.

Output

• docs/reviews/<version>/<feature-id>-analyze-report.md

Two-Phase Protocol

Phase 1 — Script Extraction

Run scripts/analyze_gate.py to extract:

• user stories
• acceptance criteria
• flows
• non-goals
• UI screens
• placeholders
• deterministic hotspot candidates

Phase 2 — Semantic Audit

Review the extracted evidence and decide whether script-flagged mismatches are:

• real gaps
• wording differences
• acceptable scope choices

Required Checks

#	Check	Pass Condition
A1	Requirement Coverage	Every user story has acceptance criteria
A2	AC Testability	Every AC is verifiable without implementation knowledge
A3	Flow Completeness	Flows have clear start/end transitions
A4	Non-goal Alignment	Requirements do not contradict non-goals
A5	UI Coverage	Every UI screen maps to one or more user stories

Output Contract (required)

The final report MUST contain a table with these columns:

Check	Script Evidence	Semantic Judgment	Result

Every A1-A5 row must be present. When upstream interface summaries are loaded, A6 must also be present.

The final report MUST also contain a dedicated ## Semantic Audit Verdict section. A raw SIGNALS_CLEAR / HOTSPOTS_FOUND script dump is not a valid gate completion artifact.

Verdicts

• PASS — all A1-A5 rows pass after semantic review, A6 passes when it applies, and ## Semantic Audit Verdict is present
• GAPS_FOUND — one or more real gaps remain
• BLOCKED — script not run, evidence missing, or row coverage incomplete

Fail-Closed Rules

• PASS is forbidden if the report only gives a summary without row-level evidence
• If script output and final judgment disagree, explain why
• If UI brief is absent, A5 must still appear as N/A with reason