Skills are free to access Create a practical AI-use policy and governance framework for HR that reflects your organisation’s risk appetite, data practices and decision-making structure. This skill guides you through acceptable use, data handling, human oversight, risk tiers and review responsibilities, so you can move from broad AI principles to workable guardrails. It is designed to work alongside existing IT and data policies, help HR teams surface the right questions, and create a policy that is ready for stakeholder review rather than a generic template.
How to install and use this skill
- Download the .skill file for the skill you want to use.
- In Claude: go to Customize → Skills, click +, then Create skill and Upload a skill. Select the .skill file and switch the new skill on.
- In ChatGPT: open Skills, select Create, then Upload from your computer, and choose the .skill file.
- Run the cold start before using the skill for real work. The first time, it will ask a short set of questions about your organisation, processes, preferences, and guardrails. Answer what you can; it will use clearly stated defaults for anything you do not know. At the end, download the updated file and use Update Skill to replace the installed version. If something changes later, open the skill’s
references/profile.md, edit the relevant answer, save it, and use Update Skill again — there is no need to complete the full interview again.
If you use Microsoft Copilot: create one focused Copilot agent per skill rather than uploading the package. In Copilot, go to Agents & Skills → New agent, then copy the skill’s SKILL.md into the agent’s Instructions field.
