agentmods.dev

writing-ai-governance-policies

writing-ai-governance-policies is a skill for Claude Code from jamesgray-ai/handsonai-plugins. It costs 217 tokens per session (3,513 once invoked), scanned A, original, MIT.

A guide for writing a company's AI governance policy: a short list of rules for AI workflows, skills, and agents, with a named check for each rule. AI governance means setting and checking how an organisation uses AI responsibly.

In plain words
What is it for?
It is for drafting, reviewing, or updating an AI policy and linking each policy rule to an actual check or clearly marked planned check.
Why use it?
It turns broad expectations into rules that one person can check quickly and avoids presenting planned checks as if they already exist. It also encourages using existing business reviews and asking before reading confidential information.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: reads .claude/ paths; mentions Claude Code; mentions Codex.

Part of the handsonai plugin — 16 skills shipped together

not rated 8repo yesterday A scan 217 tokens original MIT

Good fit It is for drafting, reviewing, or updating an AI policy and linking each policy rule to an actual check or clearly marked planned check.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies
View source ↗ jamesgray-ai/handsonai-plugins
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add jamesgray-ai/handsonai-plugins --skill writing-ai-governance-policies
Clone the repo
git clone --depth 1 https://github.com/jamesgray-ai/handsonai-plugins

Made for: Claude Code.

Or install handsonai, the plugin that ships this one along with the rest of its 16 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for writing-ai-governance-policies

README.md
[![agentmods](https://agentmods.dev/badge/skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies/github.svg)](https://agentmods.dev/skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies)
Your own site
<a href="https://agentmods.dev/skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies"><img src="https://agentmods.dev/badge/skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for writing-ai-governance-policies

Your own site · 80×15
<a href="https://agentmods.dev/skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies"><img src="https://agentmods.dev/badge/skills/jamesgray-ai/handsonai-plugins/writing-ai-governance-policies.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 217 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,513 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00217 $0.03513
Opus 5.5 $0.00087 $0.01405
Sonnet 5.5 $0.00043 $0.00703
Haiku 4.5 $0.00022 $0.00351

Measured yesterday against content hash 9441da850097, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

writing-ai-governance-policies scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/handsonai/skills/writing-ai-governance-policies/SKILL.md · 145 lines

How it starts

The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Writing AI Governance Policies

Draft and revise a short, checkable AI governance policy: one Markdown file holding a list of rules every AI workflow, skill and agent must meet, each naming the check that confirms it, plus the list of those checks. A governance record shows where one workflow stands; the policy says where every workflow should stand. The rules are built on the eight governance practices (references/starter-rules.md); the file's shape is in references/policy-format.md.

Principles

  • Every rule is checkable in minutes by one person, and names its check. Every check names the rules it covers.
  • Use the checks the business already has (a review before each release, a weekly look at the registry, the re-check dates Test sets) before proposing new ones.
  • Never show a check that doesn't exist as if it did. A check that's planned keeps a "from" note naming what will deliver it.
  • Plain words. One sentence per rule. The leader is a business leader, not an engineer.
  • Ask, never assume. Never open anything the user names as confidential, and ask before reading anything that might hold sensitive data.

Workflow

Entry

If the user wants rules for how staff use AI tools such as ChatGPT, say in one line that this skill writes rules for the company's AI workflows, skills and agents, not staff rules, and stop.

  1. Find the working folder: the folder that holds registry/ (and usually framework/). On Claude Code and Codex it's the workspace; on Claude desktop, the folder the user selected; on the web and in Microsoft 365, ai-workspace/ (or the user's own working folder) in their drive, read through the connection. If you can't read any files, ask the user to select or connect their working folder, or to paste their list of workflows, and read nothing until one of those is done. Read only inside the working folder; never open folders beside it.

    No registry/ found: say in one line that this policy covers the AI workflows in their AI Registry, so the registry comes first, and that they can set one up by saying 'set up my registry'. Stop there.

  2. Look for an existing policy: policies/ai-governance-policy.md in the working folder, or wherever the user says it lives. One exists → Review mode, below. None → First run. Never ask whether to review or write fresh when the file tells you.

Read the full file on GitHub · 145 lines

Files

What ships with it

36 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 145 lines · 217 tokens per session scan A 9441da850097

Subscribe to this mod's changes

writing-ai-governance-policies is a skill published in the GitHub repository jamesgray-ai/handsonai-plugins (8 stars, last pushed yesterday), licensed MIT. It adds 217 tokens to every session and 3,513 once invoked, about $0.0009 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-06.

Building agentsLegal and compliance
Related

writing-ai-governance-policies compared with skills from other repositories

writing-ai-governance-policies next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
writing-ai-governance-policiesjamesgray-ai/handsonai-pluginsSkill8yesterdayMITA217Claude Code
fto-triageanthropics/claude-for-legalSkill9.9k3d agoApache-2.0A81Claude Code
legal-holdanthropics/claude-for-legalSkill9.9k3d agoApache-2.0A68Claude Code
specification-writingwanshuiyin/Auto-claude-code-research-in-sleepSkill19k3d agoMITA49Claude Code
terms-of-servicegenspark-ai/genofficeSkill12k2d agoApache-2.0A32Claude Code, Codex
x-scorecardx-cmd/x-cmdSkill4.7k7d agoApache-2.0A57Claude Code, Codex
gesellschaftsrechtliche-satzungen-agbKlotzkette/claude-fuer-deutsches-rechtSkill1.8kyesterdayApache-2.0A69Claude Code