strategy-red-team
- Use it when:
- You're stress-testing a plan, challenging its assumptions or preparing a document for executive review.
- It does:
- Lists what the plan claims, keeps the load-bearing claims that would kill it if false, and attacks the strongest version of each.
- You get:
- Three to five kill-assumptions ranked by impact × likelihood × cheapness to test, each with how it fails, evidence to get this week, a kill criterion and the cheapest test.
- Also says:
- What's well reasoned and why, and what it couldn't assess.
- Unlike pre-mortem:
- That one imagines the launch already failed; this one attacks the assumptions now, while the cheapest test can still run.
- Optional:
- On request, and when another model such as Codex or Gemini is reachable, runs the plan past it and flags where they disagree.
TL;DR
From the PM Skills marketplace's pm-execution plugin. The agent plays a sharp but fair adversary. It extracts every claim the plan makes about users, market, constraints, mechanism and timeline, and separates load-bearing claims from cosmetic ones. For each load-bearing claim it first states the strongest case for it, then attacks that case, writing each failure mode as a concrete, falsifiable "Fails if ___". It ranks them by impact if wrong × likelihood of being wrong × cheapness to test, so the top of the list is what to test this week. It treats a risk as real unless the plan already cites evidence against it, but says plainly when a claim is well reasoned and never invents a weakness. Each surviving assumption gets the evidence to collect this week, the threshold at which you'd stop or change course, and the smallest experiment that would move the belief. The output is a Markdown reply, not a saved file, meant to be screenshotted, with no generic risk lists, and it ends on what to do rather than what to fear.
Summarizes the skill at 2a12295.
npx skills add https://github.com/phuryn/pm-skills --skill strategy-red-team skills.sh pageSource on GitHubAll skills from phuryn/pm-skills
Fewer weekly downloads than 79% of skills
Weekly values
| Week to | Installs |
|---|---|
| Sep 21 | 137 |
| Sep 28 | 115 |
| Oct 5 | 113 |
| Oct 12 | 98 |
| Oct 19 | 103 |
| Oct 26 | 163 |
| Nov 2 | 134 |
| Nov 9 | 111 |
Fewer downloads than 50% of skills
More stars than 70% of repos
Newer than 81% of skills
| Category |
Make the agent question you and write the plan down before it writes code. Everything for building software: planning, coding, debugging, design, frameworks, cloud, agents and skills. |
|---|---|
| Type |
A multi-step process the agent follows: interview, debug, run a TDD loop, review. |
| Effect |
Adds no side effects of its own. It shapes how the agent thinks, writes or reviews. Guidelines and routers are always effect:none, even when the task they support edits code. |
| Needs |
Instructions only. Works in any agent that supports skills. |
| Publisher |
Everyone else. Unverified vendor claims default to community. |
| Topics |
SKILL.md
From phuryn/pm-skills at 2a12295. Checked daily, last changed Nov 6, 2026.
Frontmatter
name: strategy-red-team
description: "Red-team a PRD, roadmap, or strategy by attacking its load-bearing assumptions before reality does. Steelmans then attacks each claim, ranks failure modes by impact × likelihood × cheapness-to-test, and returns the cheapest test and kill criteria for each. Use when stress-testing a plan, pressure-testing a strategy, challenging assumptions, or preparing a doc for executive review."Strategy Red-Team: Attack the Assumptions Before Reality Does
Purpose
You are a sharp, fair adversary reviewing $ARGUMENTS. Most plans only survived polite feedback. This skill finds the load-bearing assumptions that would make the plan fail, attacks them honestly, and returns — for each — the evidence to get this week, the kill criteria, and the cheapest test.
Context
A red-team is not a pre-mortem. A pre-mortem imagines the plan already failed and narrates why. A red-team attacks the load-bearing assumptions and logic now, while there's still time to test the cheapest one. It improves judgment, not just confidence.
The goal is a sharper decision, not a longer risk list. Five real kill-assumptions with tests beat twenty generic risks.
Instructions
-
Extract every claim. Read the plan and list what it asserts as true — about the user, the market, the constraint, the mechanism, the timeline. Separate load-bearing claims (if false, the plan dies) from cosmetic ones. Only load-bearing claims are worth attacking.
-
Steelman, then attack. For each load-bearing claim, first state the strongest version of why it might be true. Then attack that — not a strawman. An attack on a weak version of the claim is worthless.
-
Write each failure mode as "Fails if ___." Be concrete and falsifiable. "Fails if activation isn't actually the constraint" beats "execution risk."
-
Rank by (impact if wrong) × (likelihood wrong) × (cheapness to test). The top of the list is what to test this week — high-impact, plausibly wrong, and cheap to check. Surface that ranking; don't bury the lede.
-
Self-refute, don't fabricate. Default to "this risk is real" unless the plan already cites evidence against it. But if a claim is genuinely well-reasoned, say so plainly — a red-team that manufactures doubt is as useless as one that rubber-stamps. Never invent a weakness the plan doesn't have.
-
For each surviving kill-assumption, give the operator something to do:
- Fails if: the precise condition that breaks the plan
- Evidence to get this week: the specific data, query, or conversation that would confirm or kill it cheaply
- Kill criterion: the threshold at which you'd stop or change course
- Cheapest test: the smallest experiment that moves the belief
-
Optional cross-model mode. If the user asks for a second opinion and another model (Codex, Gemini, a second Claude) is reachable, run the same plan through it and flag where the two disagree — different model families miss different things. Default is single-model; don't add this friction unless asked.
-
Structure the output (make it screenshot-native):
## Red-Team: [plan in one line] ### Top Kill-Assumptions (ranked) For each (3–5 max): - **Claim:** [the load-bearing assertion] - **Fails if:** [concrete, falsifiable condition] - **Evidence to get this week:** [specific] - **Kill criterion:** [threshold] - **Cheapest test:** [smallest experiment] ### What's Well-Reasoned [State explicitly what holds up — and why. Don't manufacture doubt.] ### What I Couldn't Assess [Gaps where the plan didn't give enough to judge.]
Notes
- No strawmanning — attack the steelman or don't attack.
- No generic risk lists — every item must be specific to this plan.
- No fabrication — if it's sound, say so.
- Rank ruthlessly — the cheapest high-impact test is the whole point.
- The emotional job is relief from the fear of confidently shipping the wrong bet, so end with what to do, not just what to fear.