skills.rest
marcmarti9
marcmarti9COMMUNITY
Explore all skills by this author →
GitHub

adversarial-idea-review

Red-teams early product and business ideas to expose failure modes before commitment.

3|Updated Jul 28, 2026
Search Tags:#idea validation#go-to-market#product strategy#competitive analysis#business model#red team#kill gates
One-click install
npx skills add https://github.com/marcmarti9/agentit --skill adversarial-idea-review-marcmarti9
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: adversarial-idea-review
Source: https://github.com/marcmarti9/agentit/tree/main/skills/adversarial-idea-review
Command: npx skills add https://github.com/marcmarti9/agentit --skill adversarial-idea-review-marcmarti9

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Early product, feature, and business ideas often sound coherent while hiding fatal flaws in economics, adoption, competition, or operations. This Skill attacks serious candidate ideas before resources are committed, surfacing the cheapest failure modes while change is still cheap. ## Core Features & Use Cases - Structured Attack Process: A 10-step checklist covering framing, attack mapping, failure timelines, economics, adoption, competition, operations, moat analysis, kill gates, and defenses. - Bounded Verdicts: Every review ends with KILL, PIVOT, PROCEED_WITH_GATES, or PROCEED, plus the top ways the idea dies and the next cheapest test. - Evidence Discipline: Material claims are tagged as FACT, SIGNAL, ANECDOTE, HYPOTHESIS, or ASSUMPTION to prevent anecdotes from becoming market statistics. - Use Case: Before committing to a new SaaS feature, run this Skill alongside idea-refine to test pricing anchors, shadow workflows, incumbent substitutes, and support burden, then feed the verdict back into the final recommended direction. ## Quick Start Red-team this product idea before we commit to it and give me a verdict with kill gates.

Frequently Asked Questions about adversarial-idea-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I stress-test a business idea before building it?▼

Run an adversarial review that frames the candidate, maps attacks from competitors and substitutes, walks a failure timeline from first meeting to scale, and ends with a bounded verdict. Define 3-7 falsifiable kill gates so evidence can kill or change the idea early.

What is the difference between idea red-teaming and code review?▼

Idea red-teaming attacks early product, market, and business model assumptions before implementation, while code review evaluates a finished artifact or diff. Use adversarial idea review before commitment and doubt-driven-development or code-review-and-quality after implementation.

When should I use adversarial idea review with idea-refine?▼

Run it after idea-refine reduces brainstorm fragments to serious candidate directions but before convergence finishes and a Recommended Direction stands. The red-team verdict must be incorporated into the final direction, not appended as a disclaimer.

When should I not red-team an idea?▼

Skip it for trivial reversible ideas, mechanical edits, executing an already-decided plan, security threat modeling, and factual verification tasks. Red-teaming every raw brainstorm fragment wastes effort; attack only serious candidates.

How do I avoid anecdote-based decisions in product strategy?▼

Tag every material claim as FACT, SIGNAL, ANECDOTE, HYPOTHESIS, or ASSUMPTION. Never convert one review or customer story into a market-wide claim, and verify claims that affect pricing, market size, or kill decisions before treating them as facts.

Related Skills

Explore More Product & Management Skills→
MiDouTechMiDouTech

problem-statement

Create a structured problem framing narrative from user research inputs.

Official
Basic
hylarucoderhylarucoder
278

hai-idea

Score ideas across ten criteria and produce a decisive investment verdict.

Community
Intermediate
tomwangowatomwangowa
1

narrative-auditor

Extract and verify claims from articles against primary sources.

Community
Advanced
BorisserzBorisserz
1

doubt-driven-development

Runs adversarial fresh-context reviews of non-trivial outputs using doubt cycle CLAIM-EXTRACT-DOUBT-RECONCILE-STOP.

Community
Advanced