What do I need to use Policy Drafter?

The behavior being governed and why now; The real cases; The bright-line candidates; The enforcement truth.

Is Policy Drafter 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.

Policy Drafter

Q: How do I know the Policy Drafter output is good?

The skill verifies itself against its own quality checks: Every rule states its reason; Bright lines and judgment zones are visually and grammatically distinct; The worked cases are the team's real questions, reasoning shown; Enforcement describes reality — no rule survives that nobody will enforce.

🔵 Stable🕐 updated 2026-07-19 🔷 SkillSpec L3 pm-cowork

Draft an internal policy people can actually follow — the rule stated plainly with its reason, the bright lines separated from the judgment zones, the edge cases resolved by principle, and the enforcement reality stated honestly. Use when asked write our expense/remote-work/AI-use/security policy, turn this incident into a policy, our policy doc is unreadable, or people keep asking what's allowed. Produces the policy with rules-plus-reasons, the bright-line/judgment split, the worked edge cases, and the honest enforcement section.

▶ Run it free — no key needed 📝 Grade your existing draft View SKILL.md ↗

What to give it

▸The behavior being governed and why now — the incident, the ambiguity, the new tool; policies exist for reasons and the reasons belong in the text

▸The real cases — the actual questions people have asked ("can I expense the airport lounge?", "can I use AI on customer data?") — these become the worked examples, and a policy that doesn't answer them fails at its job

▸The bright-line candidates — what leadership genuinely intends as never/always, vs. what they want discretion on; drafting discovers this boundary and forces the conversation

▸The enforcement truth — what will actually happen on violation; if the answer is "probably nothing," the policy needs redesign (fewer rules, real ones), not stronger language

✅ 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.

✓Every rule states its reason

✓Bright lines and judgment zones are visually and grammatically distinct

✓The worked cases are the team's real questions, reasoning shown

✓Enforcement describes reality — no rule survives that nobody will enforce

✓The card fits a page and answers the common cases alone

⚠️ What it refuses to do

Do not draft in legalese — unparsed policies govern nothing; folklore fills the gap

Do not write "use good judgment" as a rule — name the factors and the decider or it's not policy

Do not skip the edge cases — they're the questions the policy exists to answer

Do not keep unenforceable rules for tone — each one discounts the enforceable ones

Do not publish without an owner and review date — orphan policies drift into fiction within a year

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 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="policy-drafter"></div>
<script src="https://mohitagw15856.github.io/pm-claude-skills/embed.js" async></script>

💬 Discussion

Policy Drafter is one of 750 open-source professional AI agent skills — all SkillSpec L3. Try them all in the browser · ⭐ Star on GitHub · Browse the full catalog