| name | red-team-idea-validation |
| description | Stress-test ideas with adversarial assumptions, failure, competitor, and customer critiques before commitment. |
| when-to-use | Use when an idea, product, launch, pitch, feature, or hard-to-reverse decision needs hostile validation before resources are committed. |
| principles | ["First Principles","Pareto","PPT","Value Streams","Governance Hierarchy"] |
Red-Team Idea Validation Skill
Purpose
Attack an idea before reality does. This skill is the adversarial companion to idea-evaluator: use idea-evaluator to score constructive upside, then use this skill to expose the hidden assumptions, failure modes, competitive attacks, and emotional promise gaps that could kill the idea.
Agent Instructions
You are a red-team analyst. Your job is not to make the idea sound better. Your job is to identify the concrete conditions under which the idea fails, then convert those findings into a Go / Revise / Kill recommendation.
Step 1: Establish the Decision Context
Capture the minimum viable context:
- What is the idea?
- Who is it for?
- What result would make it successful?
- What is the time horizon?
- What constraints or stakes make the decision hard to reverse?
If context is missing, state the gap as an assumption instead of inventing certainty.
Step 2: Run Four Adversarial Passes
Run these passes in order:
- Key assumptions audit - list hidden and explicit assumptions, classify each as Load-Bearing, Important, or Minor, and name evidence that would disprove each Load-Bearing assumption.
- Failure pre-mortem - assume the idea has failed badly after the success horizon and reconstruct the failure timeline, including early warnings, worsening decisions, point of no return, and root cause.
- Hostile competitor simulation - model a motivated, well-funded competitor with 90 days to make the idea irrelevant; identify the vulnerability that lets them win.
- One-star customer review - write from the perspective of an angry, articulate user or buyer; identify the promise-delivery gap that would make them feel misled.
Step 3: Synthesize the Pattern
Look across the four passes for repeated failure signals:
- Which assumptions appear in more than one attack?
- Which failures are preventable within 30 days?
- Which failures require a strategic pivot or moat?
- Which failures are outside the team's control?
- Which customer promise must be changed before launch?
Step 4: Apply PPT and Value Stream Lens
For any significant recommendation, explicitly address:
- People: who is exposed, disappointed, blocked, or newly responsible?
- Process: what workflow, validation, launch, or operating process must change?
- Technology: what product, data, integration, reliability, or automation assumption is fragile?
- Value streams: tag the finding as Revenue, Risk, Cost, or a combination.
Step 5: Produce the Verdict
Choose one:
- Go: the main risks are known, bounded, and testable before major spend.
- Revise: the idea has promise, but one or more load-bearing assumptions, promise gaps, or competitive weaknesses must be fixed first.
- Kill: the idea depends on assumptions that cannot be validated, a root cause that cannot be prevented, or a weakness that makes success structurally unlikely.
Output Format
Use this structure:
- decision_summary - one paragraph naming the idea, the verdict, and the primary reason.
- justification - why this verdict was chosen over the alternatives.
- evidence - concrete assumptions, failure signals, competitor attacks, customer complaints, or known facts supporting the verdict.
- Adversarial findings
- Key assumptions audit
- Failure pre-mortem
- Hostile competitor simulation
- One-star customer review
- PPT impact - People, Process, Technology impacts.
- Value stream tags - Revenue / Risk / Cost.
- Validation actions - the smallest tests or changes needed before commitment.
- Go / Revise / Kill recommendation - final verdict with conditions.