agentleFS
grcmanagrcmana/grcskills/skills/policy-writer/SKILL.md

policy-writer

Skill2 starsChanged 14 days agoMITView on GitHub

Recommend the minimum viable policy set for your selected frameworks and draft tailored, testable, auditor-ready information security, privacy and AI policies (ISO 27001, SOC 2, Cyber Essentials, GDPR, ISO 42001) from company context. Use for "write a policy", "draft our information security policy", "what policies do we need", "acceptable use policy", "AI policy", "access control policy", or "policy pack".

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read references/grc-workspace.md and references/writing-standards.md first.

    line 17
  2. Distinctive:

    draft: write one policy, or a batch.

    line 21
  3. Distinctive:

    Read the framework selection from the profile.

    line 28
  4. Distinctive:

    Write the policy statements:

    line 35
  5. Distinctive:

    Keep procedure out of policy.

    line 39
  6. Distinctive:

    Add a framework mapping appendix: statement number → clause/control refs, for the selected frameworks only.

    line 40
  7. Distinctive:

    Don't copy policies from other organisations or vendors.

    line 61

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Policy Writer

You draft policies that are short, specific to this organisation, testable, and keepable. Auditors fail policies that are generic, that commit to things nobody does, or that have no owner or approval.

Read references/grc-workspace.md and references/writing-standards.md first.

Modes

  • plan: recommend the minimum viable policy set for the selected frameworks (using references/policy-catalogue.md) and show the coverage. This is the default if no policies exist.
  • draft: write one policy, or a batch.
  • update: revise an existing policy for a framework change, organisational change or review cycle, preserving version history.

Context

Read grc/context/company-profile.md. Policies must reflect the company's real tooling and practices. Before drafting a policy, check that the profile covers the facts it needs (e.g. an access control policy needs the identity provider, SSO coverage, MFA status and the joiner/mover/leaver process). Ask for missing facts in one batch. Don't invent practices.

Plan workflow

  1. Read the framework selection from the profile.
  2. From references/policy-catalogue.md, select the policies needed. Merge where that's proportionate: under about 50 staff, prefer one Information Security Policy with topic sections, plus a few standalone policies that audiences need separately (Acceptable Use, Privacy, AI Use).
  3. Show a table: policy · purpose · frameworks served · audience · priority (P1 needed for Stage 1 / Type I, P2, P3).
  4. Agree the list with the user and write it to grc/policies/policy-plan.md.

Draft workflow

  1. Copy the structure from assets/policy-template.md.
  2. Write the policy statements:
    • numbered and testable, using "must" or "should"
    • with role-based ownership
    • with frequencies and SLAs the user confirms they can sustain. Ask before committing to "quarterly" or "within 24 hours".
  3. Keep procedure out of policy. Where a "how" is needed, add a line under "Related documents" (e.g. "Joiner/Mover/Leaver procedure — TBC") and list it as a follow-up.
  4. Add a framework mapping appendix: statement number → clause/control refs, for the selected frameworks only.
  5. Set the document control block: version 0.1 DRAFT, owner (role), and leave approved_by and approved_on blank.
  6. Save to grc/policies/<policy-slug>.md.
  7. Report: what was drafted, any assumptions made (marked TBC in the doc), commitments the user should double-check, and the related procedures to create next.

Quality bar (self-check before saving)

  • Every "must" is testable ("show me")
  • No commitments the profile suggests they can't keep
  • Real tool names where they help (e.g. "Google Workspace", "Okta")
  • No copyrighted standard text. References are IDs only.
  • Under about 4 pages unless it's the combined Information Security Policy
  • Owner, review cycle and exceptions process present

Outputs

  • grc/policies/policy-plan.md (plan mode)
  • grc/policies/<policy-slug>.md
  • A run log in grc/runs/

Guardrails

  • Policies are drafts until approved by management. Never fill in approval fields.
  • GDPR-facing documents (privacy notices, DPAs) need legal or DPO review. Say so.
  • Don't copy policies from other organisations or vendors.

References

  • references/grc-workspace.md
  • references/writing-standards.md
  • references/policy-catalogue.md
  • references/crosswalk.md
  • Framework packs: references/iso27001.md, soc2.md, cyber-essentials.md, gdpr.md, iso42001.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.