What do I need to use Devil's Twin?

The document; Who would oppose this in real life (optional).

Is Devil's Twin free to use?

Yes — run it free in the browser playground (no API key needed for the sponsored trial, or bring any provider key), or install it into Claude Code, Cursor, ChatGPT and more via npx pm-claude-skills. MIT-licensed and open source.

Devil's Twin

Q: How do I know the Devil's Twin output is good?

The skill verifies itself against its own quality checks: The memo argues FROM the opponent's premises — deleting "not" from your claims would not reconstruct it; It concedes at least one of your points — full-spectrum opposition is a strawman wearing a suit; At least one of your claims is marked 💀 or the twin explains why your case is unusually airtight (rare; suspicious); The pre-emption paragraph is drop-in ready — your voice, your document's structure, no "as some may argue" throat-clearing.

🔵 Stable🕐 updated 2026-10-06 ✅ 4.8/5 🔷 SkillSpec L3 pm-warroom

The strongest possible case AGAINST what you just wrote — argued to win, not to check a box. Use when a document is about to ship and everyone around it already agrees: the twin writes the opposition's best memo (not a critique of yours), so you meet the real counter-argument before your audience does. Produces the opposing memo, the map of which of your claims it defeats/dents/leaves standing, and the pre-emption paragraph worth adding.

▶ Run it free — no key needed 📝 Grade your existing draft ⬇ Add to Claude (works in Duet) 📱 Use as an app View SKILL.md ↗ What changed

🗣 Say this to your agent

“Argue against this as hard as you can.”“Write the opposition's best memo.”“Everyone agrees with this doc: what's the strongest case against it?”“Make the case that we should not do this.”

What to give it

▸The document — full text. The twin argues against the strongest version of what you wrote, so it must see all of it.

▸Who would oppose this in real life (optional) — (optional but sharpening) — the CFO, the incumbent team, the sceptical customer, the regulator. The twin adopts their premises, not a generic contrarian's.

✅ The bar it holds itself to

Every skill in this library self-verifies — these are this skill's own quality checks, straight from its definition.

✓The memo argues FROM the opponent's premises — deleting "not" from your claims would not reconstruct it

✓It concedes at least one of your points — full-spectrum opposition is a strawman wearing a suit

✓At least one of your claims is marked 💀 or the twin explains why your case is unusually airtight (rare; suspicious)

✓The pre-emption paragraph is drop-in ready — your voice, your document's structure, no "as some may argue" throat-clearing

✓If the twin's case is stronger overall, the verdict says so plainly

⚠️ What it refuses to do

Do not write a critique with quotations — the deliverable is the opposition's own memo, structure and all

Do not make the twin stupid to make you feel good — a weak twin is worse than none; it inoculates you against the wrong argument

Do not have the twin invent facts — it may reinterpret your evidence and add commonly-known context, never fabricate data

Do not skip the verdict to stay diplomatic — "repair first" beats a polite shrug

Do not use the twin on documents whose audience is hostile already — it's for consensus rooms, where nobody else will say this

Install

npx pm-claude-skills add --agent claude   # or codex · cursor · gemini · hermes
# or one-line MCP (every skill, any client):
claude mcp add pm-skills -- npx -y -p pm-claude-skills pm-claude-skills-mcp

Related skills

🔌 Embed this skill

Drop this on your blog, docs, or site — it renders a "Run this skill" card:

<div data-pm-skill="devils-twin"></div>
<script src="https://mohitagw15856.github.io/pm-claude-skills/embed.js" async></script>

💬 Discussion

Devil's Twin is one of 1326 open-source professional AI agent skills — all SkillSpec L3. Try them all in the browser · ⭐ Star on GitHub · Browse the full catalog