Product-lens reviewer
SkillDocs & knowledgeReviews any document — strategy doc, spec, brief, launch plan, proposal — through a product POV across 6 dimensions: target-problem clarity, persona focus, metric quality (anti-vanity), differentiation defensibility, scope discipline, and shipping-as-experiment framing. Produces a scored product-review-report with FAIL/WARN/PASS verdicts and rewrite suggestions. Triggers: "product review", "review through product lens", "PM review", "is this strategy any good". Recommended upstream: strategy-doc, positioning, icp-research. Run before locking strategy docs, sending client proposals, or shipping launch plans. Composes with voice-reviewer and design-reviewer. NOT for code/PR review (engineering:code-review).
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Product-lens reviewer skill
What this skill tells your AI
The instructions your AI receives, as published by matteotitta/genesys-skills in skills/meta/catalog/product-lens-reviewer/SKILL.md and read by ahel’s review.
Review any document through a product POV. Composes with voice-reviewer (style/voice) and design-reviewer (visual/UX). Adapted from ce-product-lens-reviewer agent in EveryInc/compound-engineering-plugin v3.5.0 (MIT).
This is the structural counterpart to voice-reviewer. Voice asks "does this sound right?"; product asks "does this think right?"
When to run
Invoke when the user says:
- "Run product review on [doc]"
- "Review this through a product lens"
- "PM review this strategy"
- "Is this strategy any good?"
- "Pre-lock review on [strategy-doc / proposal / launch plan]"
Do NOT invoke when:
- User wants voice review →
voice-reviewer - User wants visual review →
design-reviewer - User wants code review →
engineering:code-review
Composes with: voice-reviewer, design-reviewer, scope-guardian-reviewer (often run together as a review-pass family before locking strategic docs).
Inputs
Required:
- The document to review (text)
Recommended:
strategy-doc(if reviewing a strategy doc, comparing against itself; if reviewing a downstream artifact like a launch plan, comparing against the upstream strategy)positioning(cross-check differentiation defensibility)icp-research(cross-check persona focus)
The 6 dimensions
Each dimension scores PASS / WARN / FAIL with quoted evidence.
| # | Dimension | Trigger |
|---|---|---|
| 1 | Target-problem clarity | Is the problem specific, recurring, expensive? Cited evidence vs. generic? |
| 2 | Persona focus | One persona ideally; if 3+ — FAIL (audience, not persona) |
| 3 | Metric quality | SMART criteria; anti-vanity-metrics rule — page views/impressions/MAU without conversion → FAIL |
| 4 | Differentiation defensibility | Would a skeptical buyer choose this over alternatives? Cite the alternative + why this beats it |
| 5 | Scope discipline | Does the doc stay in its lane? Strategy doesn't include requirements (K6); spec doesn't include strategy; proposal doesn't blur with positioning |
| 6 | Shipping-as-experiment framing | Is the ship treated as data generation (K1)? Hypothesis stated? Falsifiable? |
For full rule set with examples → the premium reference (deferred — created on first real run).
Verdict logic
| Inputs | Verdict |
|---|---|
| All PASS | Ship it |
| Any WARN, no FAIL | Minor fixes recommended |
| Any FAIL | Fix before shipping |
Anti-hallucination guardrails
- Quote the actual text when flagging an issue. No paraphrase.
- Don't flag style preferences as violations. This is product, not voice. If the issue is tone, route to voice-reviewer.
- Don't over-flag intentional choices. A two-persona strategy might be intentional — check the doc's justification before flagging.
- Be specific about fixes. "Sharpen the persona" is not a fix. "Drop 'marketers' — keep only 'product managers in B2B SaaS, 50-200 employees, post-Series A'" is a fix.
Self-roast (pre-delivery)
- All 6 dimensions evaluated
- Each finding has quoted evidence
- Fixes are specific enough to implement without guessing
- Composition checked — voice/design issues routed to the right reviewer (don't report style issues here)
- At least one cross-reference to upstream artifact (strategy-doc / positioning / icp-research) when available
Product-lens review: {doc name} ({YYYY-MM-DD})
Verdict: Ship it | Minor fixes | Fix before shipping
Score table
| Dimension | Verdict | Notes |
|---|---|---|
| 1. Target-problem clarity | PASS/WARN/FAIL | ... |
| ... | ... | ... |
Findings
§{N} {Dimension} — {verdict}
"{quoted text}" Fix: {specific replacement or reframe}
Cross-references checked
- {strategy-doc / positioning / icp-research} — {alignment notes}
## Composition rule reference
Product-lens-reviewer is one node in the **lens-reviewer pattern** (P6). Composes with voice-reviewer (style), design-reviewer (visual), scope-guardian-reviewer (scope creep), coherence-reviewer (internal consistency, when built). See [.claude/rules/pm-loop.md](../../../../rules/pm-loop.md).
## Attribution
Adapted from [EveryInc/compound-engineering-plugin](https://github.com/EveryInc/compound-engineering-plugin) v3.5.0 (MIT). Source agent: `ce-product-lens-reviewer`. Pattern: pluggable lens-reviewers applied to strategy docs (P6 from /steal Phase 4).
Signals
- GitHub stars
- 36
- Forks
- 14
- Last commit
- Jul 2026
Advanced
- Catalog kind
- skill
- Gateway key
product-lens-reviewer- Source
- github.com/matteotitta/genesys-skills