← Skills Library
Agent Teams

Research Team

Spawn a 3-agent research team (Researcher + Strategist + Critic) for market research, competitive intelligence, and pre-PRD discovery.

How It Works

Research Team · Workflow
Researcher → Strategist → Critic, then a calibrated brief.
▸TriggerResearch question / topic · Market intel, competitive scan, or pre-PRD discovery
1
Frame the question
Decision it informs · scope · output format · time box
2
Researcher gathers live data
Web search · pain points · 5-10 sourced data points — no interpretation
3
Strategist synthesizes
Top insights · maps to context · recommendation + risks
4
Critic stress-tests
Steelman · attack weakest claims · confidence 1-10 per claim
✓Final research brief · TL;DR · recommendation · sourced evidence · confidence High/Med/Low
  1. Research question / topic: Market intel, competitive scan, or pre-PRD discovery
  2. Frame the question: Decision it informs · scope · output format · time box
  3. Researcher gathers live data: Web search · pain points · 5-10 sourced data points — no interpretation
  4. Strategist synthesizes: Top insights · maps to context · recommendation + risks
  5. Critic stress-tests: Steelman · attack weakest claims · confidence 1-10 per claim
  6. Final research brief: TL;DR · recommendation · sourced evidence · confidence High/Med/Low
ↆ download card

Invocation Triggers

/research-teammarket researchcompetitive intelpre-PRD research

Use Cases

  • Market research before writing a PRD for a new feature
  • Competitive intelligence on a rival product or category
  • Validate assumptions before committing engineering resources

The Problem

PRDs get written in a vacuum. By the time you discover the space is crowded, or that a competitor already solved this problem, you have committed two sprints of engineering time. Research should inform the decision to build — not justify it after you have already decided. Most teams skip market research because it feels slow. Research Team runs in one command and produces a calibrated brief in under an hour.

What It Does

  1. 1
    Frame the research question

    Define what decision this research informs, what is in and out of scope, and what format the output should take. Bad research questions produce bad research. "Should we build X?" is better than "tell me about X."

  2. 2
    Researcher gathers live data

    Uses web search and fetch to collect current information — not training data. Market landscape, competitive positioning, user pain points from forums and reviews, concrete data points with sources. Gathers without interpreting.

  3. 3
    Strategist synthesizes

    Takes the raw research and maps it to your specific context. Identifies the 3-5 most important insights, the opportunity or whitespace, and the recommendation. Articulates the risks.

  4. 4
    Critic stress-tests

    Steelmans the strategy, then attacks the weakest claims. Identifies blind spots, challenges assumptions, and rates each key claim 1-10 for confidence. Does not propose alternatives — only critiques the existing recommendation.

  5. 5
    Final research brief

    Synthesizes all three outputs into a structured brief: TL;DR, recommendation with confidence rating, key evidence with sources, risks and open questions. Human-readable and decision-ready.

What You Get / What It Doesn't Do

What you get
  • Research brief with TL;DR and explicit recommendation
  • Confidence rating per key claim (1-10, from the Critic)
  • Market landscape with sourced, current data points
  • Competitive analysis with positioning gaps
  • Top risks and open questions that should be resolved before committing to build
What it doesn't do
  • Talk to real users — the Researcher reads what users say, not what they tell you
  • Validate revenue assumptions or market size beyond what is publicly available
  • Replace a proper discovery sprint for genuinely novel or high-stakes decisions
  • Make the product decision — it informs the decision, you make it

Tips

Confidence ratings are the most valuable output

A key claim rated 4/10 by the Critic is a risk. If your recommendation depends on it, that is where to do more research before committing.

Use --pre-prd flag first

If you're not sure whether an idea is worth researching, run /pressure-test first. Research Team goes deeper — use it when the idea has already survived basic scrutiny.

The brief should answer one question

Would we still build this if a competitor launched tomorrow? If the brief doesn't answer this, it hasn't done its job.

Get the Skill

Pro SkillPRO

Unlock the full Research Team SKILL.md — drop it into ~/.claude/skills/ and trigger it by name.

What you unlock
  • Research brief with TL;DR and explicit recommendation
  • Confidence rating per key claim (1-10, from the Critic)
  • Market landscape with sourced, current data points
  • Competitive analysis with positioning gaps
...

Commonly Used With

IntelligencePressure TestStress-tests a feature idea against constraints, users, and your existing system in one pass. Verdict: kill or proceed with conditions. No building required.IntelligencePRD WriterIdea to structured PRD in one command. Scope, requirements, measurable success criteria, and the open questions that will block engineering if you don't answer them first.
Skills Library

29 more production skills ready to install.

Browse All Skills