> ## Documentation Index
> Fetch the complete documentation index at: https://docs.encoreos.io/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Skill Governance — Approval & Version History

> Review and approve org-authored AI skills before they can drive AI, and inspect the immutable version history with one-click rollback.

# AI Skill Governance — Approval & Version History

> Audience: reviewers who approve org-authored AI skills.
> Spec: `specs/pf/specs/PF-120-org-authored-ai-skill-governance.md`.

Governance covers the lifecycle of the org-authored and org-customized AI skills the
[Create Custom Skill wizard](./ai-custom-skill-admin-guide.md) produces: an **approval
queue** where a reviewer signs off before a skill can drive AI, and an **immutable version
history** with one-click rollback. It is the reviewer/approver companion to the authoring
guide.

## Approval queue

Reach the queue at `/settings/ai-skills/approvals` (heading **AI Skill Approvals**), guarded
by the `pf.ai_skills.approve` permission. Regulated categories (clinical, compliance, HR,
housing) require **second-person** approval — a `pf.ai_skills.approve` holder who is **not**
the submitter — so a skill cannot drive AI until someone else signs off. Non-regulated skills
self-activate at submission; imported skills route to review with a mandatory sensitivity
class and a no-PHI attestation.

<Frame caption="The AI Skill Approvals queue: the regulated “Clinical Intake Summarizer” carries a Regulated badge and its Approve button is disabled by the self-approval guard (the admin submitted it), while the imported, non-regulated “Imported Ops Helper” carries an Imported badge and its Approve button is enabled.">
  <img src="https://mintcdn.com/encoreos/9uGfYOeeHogjmoRk/images/pf/pf-120-approval-queue.png?fit=max&auto=format&n=9uGfYOeeHogjmoRk&q=85&s=0c23cd4c55789f6ea18d4bd16cbdc86c" alt="AI Skill Approvals queue listing two org-authored skills awaiting review — one regulated with Approve disabled, one imported with Approve enabled" width="1440" height="900" data-path="images/pf/pf-120-approval-queue.png" />
</Frame>

## Version history and rollback

Every create and every prompt/config edit inserts an **immutable** snapshot into
`pf_ai_skill_versions` (insert-only — no update or delete). Open the version-history drawer
from a skill's **Configure** dialog on `/settings/ai-skills`. **Rollback** writes a *new
forward* version capturing the rolled-back state — history is never rewound — and a
rolled-back regulated skill re-enters review and needs fresh approval.

<Frame caption="The immutable Version History drawer for the regulated skill, newest-first: v3 (marked Current — the rollback-forward version), v2, and v1, each with its change reason, author, and timestamp, plus a “Roll back to this version” action.">
  <img src="https://mintcdn.com/encoreos/9uGfYOeeHogjmoRk/images/pf/pf-120-version-history.png?fit=max&auto=format&n=9uGfYOeeHogjmoRk&q=85&s=c12f7984d90b672550f90dcc70510f84" alt="Version History drawer showing the immutable v1 to v2 to rollback-forward v3 snapshot chain with v3 marked Current" width="1440" height="900" data-path="images/pf/pf-120-version-history.png" />
</Frame>

## What gets audited

Every lifecycle transition emits a platform audit entry — `pf.ai_skill.submitted`,
`pf.ai_skill.approved`, `pf.ai_skill.rejected`, `pf.ai_skill.rolled_back` — with actor, org,
skill id, and from/to state. Write access gates on `pf.ai_skills.manage` and approve/reject
on `pf.ai_skills.approve`, so the UI controls and the row-level security agree.

## See also

<Columns cols={2}>
  <Card title="AI Custom Skill Authoring" icon="wand-magic-sparkles" href="/platform/ai-custom-skill-admin-guide">
    Author a net-new organization AI skill.
  </Card>

  <Card title="AI Skills" icon="sparkles" href="/pf/ai-skills">
    Browse, configure, and manage AI skills.
  </Card>
</Columns>