agentsclimarketplace

Pm competitor analysis

Skill sparktech-ai/skills/skills/pm-competitor-analysis

Use when user wants deep competitor feature analysis. Triggers: "analyze competitor features", "teardown competitors", "what features do competitors have", "competitor deep dive". Requires knowing which competitors to analyze—if unclear, suggest market-research first.From its SKILL.md

Install
npx -y skills add sparktech-ai/skills --skill pm-competitor-analysis

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.2 KB, as published. Nobody here has run it

Competitor Teardown

Deep dive into competitor features to inform product design decisions.

When to Use

  • Have identified competitors and want to understand their features in depth
  • Need to find differentiation opportunities before writing PRD
  • Want to understand what users love/hate about existing solutions

What This Skill Does

Analyzes competitors on three dimensions:

  • Core Features - What functionality do they offer
  • Highlight Features - What users love, differentiators
  • Pain Points - What users complain about, missing features

Process

Step 1: Identify Competitors

Same-session handoff: If user just completed /pm-market-research in this conversation, use the competitors identified there directly.

New session: Check for docs/prd/research-*.md files first:

  • Research exists → Ask: "Found research for [topic] with competitors [list]. Analyze these?"
  • No research → Ask: "Do you have specific competitors in mind, or should we use /pm-market-research to identify them first?"
    • User has competitors → Proceed with user-specified list (2-5 competitors)
    • User unsure → Guide to /pm-market-research first

Step 2: Analyze Each Competitor

For each competitor, gather both features and sentiment in parallel:

Features (via WebFetch):

  • Official website, docs, feature pages
  • Build feature inventory: name, description, category

User Sentiment (via WebSearch):

What to FindSearch Examples
Positive reviews"[competitor] review", "[competitor] best feature"
Comparisons"[competitor] vs", "[competitor] alternative"
Complaints"[competitor] frustrating" reddit, "[competitor] missing feature"
Feature requests"[competitor] wish it had", "[competitor] feature request"

Extract:

  • Highlight features: Features users explicitly praise
  • Pain points: Features users complain about or wish existed

Complete one competitor fully before moving to the next.

Step 3: Output Document

Generate docs/prd/competitor-<topic>.md:

# <Topic> Competitor Teardown

## Analysis Scope
- **Purpose**: [Why this analysis]
- **Competitors**: [List of analyzed competitors]

## Competitor Breakdown

### [Competitor A]

**Core Features:**
| Feature | Description | Notes |
|---------|-------------|-------|
| ... | ... | ... |

**Highlight Features:**
- [Feature]: [Why users love it, source]

**Pain Points:**
- [Feature/Gap]: [User complaints, source]

### [Competitor B]
...

## Feature Comparison Matrix
| Feature | Competitor A | Competitor B | Competitor C | Opportunity |
|---------|--------------|--------------|--------------|-------------|
| ... | ✅/❌/⚠️ | ✅/❌/⚠️ | ✅/❌/⚠️ | ... |

Legend: ✅ Well implemented | ⚠️ Partially/poorly implemented | ❌ Missing

## Insights

### Common Weaknesses
- [What all competitors do poorly or miss]

### Differentiation Opportunities
- [Where a new product could win]

### Table Stakes
- [Features that must exist to compete]

After generating the document, suggest: "Ready to define your product? Use /pm-prd to create a PRD based on this analysis."

Key Principles

  • Deep over broad: Focus on feature-level detail, not market-level analysis
  • User voice matters: Prioritize real user feedback over marketing claims
  • Actionable insights: Every finding should inform product decisions
  • Match user's language: Output in the same language the user uses in conversation

Handling Edge Cases

  • Competitor has no public info: Note the gap, focus on user reviews and community discussions instead.
  • No user feedback found: Note as "Limited user feedback available", rely more on feature comparison.
  • Too many competitors: Recommend focusing on top 3-5 most relevant ones.
  • User abandons mid-process: Save partial analysis to draft file for later resumption.

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.