Community

competitor-briefs

Open Agent Skill · research · works with Claude Code, Codex, Cursor, and compatible agents

Research competitors into structured comparison briefs with evidence and gaps. Use whenever the user asks for competitive analysis, competitor research, market landscape, battlecards, or how we compare to another product — even if they only name one rival.

View on GitHub

Competitor Briefs

Good competitive work is sourced, dated, and decision-oriented. Separate facts from inference. Unknowns beat confident fiction.

Workflow

  1. Clarify the decision: pricing page, sales battlecard, build-vs-buy, roadmap gap, investor landscape.
  2. Fix the comparison set (2–5 rivals + your product if relevant).
  3. Collect dimensions that matter: ICP, core jobs-to-be-done, pricing model, key features, integrations, packaging, proof (customers, compliance).
  4. For each claim, attach evidence (URL, doc, quote) or mark unverified.
  5. Synthesize: where you win, lose, and what to learn — not a feature laundry list without prioritization.

Output format

# Competitor brief: <focus>
**Date:** …
**Decision this supports:** …
**Sources checked:** … (list)

## Snapshot table
| Dimension | Us | Comp A | Comp B |
|---|---|---|---|

## Competitor profiles
### <Name>
- Positioning:
- ICP:
- Pricing (if public):
- Strengths:
- Weaknesses / gaps:
- Notable proof:
- Sources:

## Implications
- Product:
- GTM / sales:
- Risks if we ignore:

## Open questions
- …

## Confidence
high | medium | low — and why (source quality, recency)

Rules

  1. Never invent pricing, customer logos, or feature support. Label estimates.
  2. Prefer primary sources (vendor site, docs, pricing page, SEC filings) over random blogs; note access date when possible.
  3. Compare on jobs-to-be-done, not raw feature counts.
  4. Call out when a "competitor" is adjacent (different ICP) vs direct.
  5. If offline / no network, produce a research plan and interview questions instead of fake findings.
  6. Keep sales battlecards honest; overclaiming destroys trust in the field.

Edge cases

  • Single competitor deep-dive: still include a small "alternatives" section so the market is not a false duel.
  • Crowded market: cluster by archetype (open source, enterprise suite, point tool) before naming every vendor.
  • Fast-changing pricing: screenshot-level caution; mark "verify live".
  • User wants only a matrix: deliver the table plus a 5-line implication summary minimum.