AgentSkillsCN

product-manager

产品管理：PRD、RICE优先级排序、指标体系。

SKILL.md
--- frontmatter
name: product-manager
description: Product management: PRDs, RICE prioritization, metrics
allowed-tools: Read, Write, Grep

Product Manager

Assists with core product management workflows including research synthesis, requirement documentation, feature prioritization, and strategic communication.

When to Use

  • •Analyzing user interviews, surveys, or feedback
  • •Writing or reviewing PRDs and product requirements
  • •Prioritizing features or roadmap items
  • •Preparing stakeholder updates or presentations
  • •Interpreting product metrics and analytics
  • •Conducting competitive analysis

Instructions

User Research Analysis

When analyzing user research:

  1. •Read the provided transcripts, feedback, or survey data
  2. •Identify the top 3-5 pain points with supporting quotes
  3. •Group insights by themes (not by individual users)
  4. •Prioritize by frequency AND impact
  5. •Suggest 2-3 actionable opportunity areas
  6. •Include specific quotes to support each finding

Feature Prioritization

When prioritizing features, use the RICE framework:

  • •Reach: How many users impacted per time period?
  • •Impact: Confidence score (0.25=minimal, 0.5=low, 1=medium, 2=high, 3=massive)
  • •Confidence: Data quality (0-100%)
  • •Effort: Person-months required
  • •RICE Score = (Reach × Impact × Confidence) / Effort

Present results in a table with reasoning for each score.

PRD Writing

When creating or reviewing PRDs, ensure these sections:

  1. •Problem Statement: Clear user problem with evidence
  2. •Success Metrics: Quantifiable measures (not "improve UX")
  3. •User Stories: Format: "As a [user], I want [action] so that [benefit]"
  4. •Acceptance Criteria: Testable, specific conditions
  5. •Edge Cases: Error states, boundary conditions, empty states
  6. •Out of Scope: What we're explicitly NOT building

Flag any missing or unclear sections.

Stakeholder Communication

When drafting communications:

  1. •Lead with impact/outcome, not features
  2. •Match tone to audience (C-level: business impact; Eng: technical details)
  3. •Use specific metrics, not vague terms like "better" or "improved"
  4. •Include next steps with owners and timelines
  5. •Be transparent about blockers/challenges

Metrics Analysis

When analyzing metrics:

  1. •Calculate key ratios (DAU/MAU, retention cohorts, conversion rates)
  2. •Identify trends (week-over-week, month-over-month)
  3. •Flag anomalies requiring investigation
  4. •Distinguish between correlation and causation
  5. •Provide 2-3 actionable recommendations

Quick Reference

Problem Validation Checklist:

  • • Problem clearly articulated?
  • • Validated with real users (not assumptions)?
  • • Frequent/painful enough to solve?
  • • Users will pay/engage more if solved?
  • • Technically feasible within constraints?

Go/No-Go Decision Criteria:

  • •Strategic alignment with company vision?
  • •Solves a validated user problem?
  • •Moves key metrics meaningfully?
  • •Team can build and maintain it?
  • •Strengthens competitive differentiation?

Guidelines

  • •Focus on outcomes (metrics moved) over outputs (features shipped)
  • •Validate assumptions with data before building
  • •Write testable, specific requirements (avoid "intuitive" or "easy to use")
  • •Consider edge cases: errors, empty states, loading states
  • •Question vanity metrics (prioritize engagement over page views)
  • •Be explicit about trade-offs in prioritization decisions

Automatic Triggers:

  • •User pastes interview transcripts or feedback
  • •User asks to prioritize features or compare options
  • •User mentions "PRD", "product requirements", or "user stories"
  • •User shares metrics data or analytics
  • •User requests stakeholder updates or presentations