/five-lens-review

name: five-lens-review
description: "Evaluate a product-direction question through five expert lenses, Senior PM, Senior Product Engineer, Senior Product Designer, Senior Design Engineer, and a realistic end user, then synthesize the conclusions into a single recommendation. Use whenever a user-facing decision has multiple competing constraints (lifecycle behavior, data-model trade-offs, destructive actions, UX choices that touch retention) and a one-perspective answer would miss something. Trigger phrases: 'what should we do about X', 'what happens when the user Y', 'how should we handle this case', 'I need to think through this', 'help me decide', 'we don't have answers for these questions', 'spawn personas', 'evaluate from all POVs'. Especially use this when the user is mid-implementation and surfaces an edge case the spec didn't cover, that's when single-perspective reasoning silently locks in the wrong default."

Five-Lens Review

Spawn five specialized subagents in parallel, Senior PM, Senior Product Engineer, Senior Product Designer, Senior Design Engineer, and a realistic end user, each examining the same product-direction question through their own discipline's lens. Then synthesize their convergent themes, surface their disagreements, and produce a single actionable recommendation.

The output is a structured memo the user can decide against, not a slide deck of opinions.

---

When this skill is the right tool

Use it when ALL of these are true:

• There's a **user-facing decision** at stake (not a pure refactor or internal optimization)
• The decision has **multiple competing constraints** (data integrity vs UX simplicity, retention vs implementation cost, spec-fidelity vs user reality)
• A single-perspective answer would **silently lock in a bad default** that's expensive to reverse later
• The user has surfaced an edge case the spec didn't cover, OR is asking "what should we do about X"

Skip this skill for:

• Pure implementation questions ("how do I refactor this loop"), that's not a product question
• Decisions where the user already has a clear preference and just wants execution
• Bug fixes with one obvious correct answer
• Anything where running five parallel subagents would be over-engineering the question

The right test: would the user feel like the decision benefits from being interrogated from five disciplines? If yes, use this. If you're not sure, ask.

---

Step 1: Frame the decision precisely

Before spawning anything, write a one-paragraph **decision context** that every persona will receive. The personas have no session memory; the brief is the only thing they see. It must contain:

• **What is the product/feature?** (3-5 sentences of background)
• **What is the current behavior?** (what code/design does today, with file paths or specifics)
• **What is the question or edge case?** (the literal thing being decided)
• **What constraints already exist?** (spec rules, voice guidelines, architectural patterns, deadlines)

If the question is ambiguous, ask the user to clarify before spawning. A vague brief produces vague memos that don't reconcile.

---

Step 2: Spawn five subagents in parallel

In a **single message**, dispatch five subagent calls using the `Agent` tool (or `Task` tool / whatever the runtime exposes) with `model: "sonnet"`. All five receive the **same decision context** plus a persona-specific lens. They run concurrently and notify on completion.

Each persona's brief should follow this shape:

Adopt the lens of a [PERSONA]. Your job is to [PERSONA-SPECIFIC GOAL].

## Context (you have no session memory)
[The shared decision context from Step 1]

## The questions to think through
[Enumerated questions, 3-7 of them, framed for this persona's discipline]

## Your [PERSONA]-specific lens
[3-6 bullets pointing at what this persona uniquely contributes, see Step 3]

## Deliverable
[Structured memo shape, see Step 4. ~700-900 words.]

End with: "[A one-line takeaway prompt the persona must answer]"

The five lenses, with what each uniquely contributes:

Senior Product Manager

**Goal**: Surface the right questions, articulate trade-offs, recommend a decision framework from product-strategy lens. Do NOT write code or get into implementation detail. **Unique contribution**: Which decisions are user-retention critical vs nice-to-have. What's the cheapest v0 answer + the eventual right answer. Which choices ENABLE future features vs LOCK US OUT. GDPR / data-retention concerns. What must decide NOW vs can defer. **End with**: "If we don't decide X, we'll regret it because Y."

Senior Product Engineer

**Goal**: Identify technical implications, data-model trade-offs, migration paths, edge cases. Get specific about implementation consequences of different decisions. **Unique contribution**: Which decisions create migration debt later. Where query complexity explodes. What testing gaps each decision creates. Indexing implications. Cheapest implementation that doesn't paint into a corner. Schema additions (add now vs defer). **End with**: "The one schema change I'd push to land THIS phase to avoid pain later is X."

Senior Product Designer

**Goal**: Design the user's mental model, surface where current behavior contradicts intuition, recommend microcopy + visual decisions. **Unique contribution**: Where the user's mental model BREAKS with current behavior. Microcopy needed at each transition. IA decisions, new sections/views needed. The right level of nuance (surface everything vs hide complexity). Modal/drawer patterns for the explanatory moments. Voice-of-product alignment. **End with**: "The one design decision I'd make NOW to prevent the most confused-user moments later is X."

Senior Design Engineer

**Goal**: Recommend how to IMPLEMENT design decisions in concrete component patterns, state machines, and reactive UI that hold up across the feature's lifecycle. **Unique contribution**: Component-pattern recommendations