Skip to content
Back to skills

Pui Govern

ASecurity

Audit one bounded Proto UI collaboration-governance slice and route policy-determined drift toward reversible correction. Use across Issues, pull requests, Discussions, labels, milestones, Project, rules, permissions, or Actions. This evidence leaf is read-only; authorized corrections continue in their registered leaves.

  • 36 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 4, 2026
designgogitapisecuritydocumentation

Works with

  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned October 4, 2026

npx -y skills add Proto-UI/Proto-UI --skill pui-govern --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Pui Govern?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Pui Govern
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/proto-ui-pui-govern/badge)](https://www.skillsdirectory.com/skills/proto-ui-pui-govern)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: pui-govern
description: Audit one bounded Proto UI collaboration-governance slice and route policy-determined drift toward reversible correction. Use across Issues, pull requests, Discussions, labels, milestones, Project, rules, permissions, or Actions. This evidence leaf is read-only; authorized corrections continue in their registered leaves.
---

# Audit collaboration state

1. Require a current capability envelope and a bounded set of repositories, surfaces, fields, and observation time.
2. Read collaboration policy and the latest applicable dated record. Keep stable policy separate from current observation.
3. Query every in-scope surface with pagination and record API scopes, filters, truncation, and unavailable fields.
4. Assign one meaning to each surface: conversation, bounded work, operational state, search property, program outcome, integration unit, review, machine evidence, or delivery fact.
5. Detect duplicate taxonomy, stale states, conflicting owners, unprotected expectations, missing controls, and projection drift. When current policy fixes the target state, identify the exact reversible correction and eligible registered transition instead of escalating it as a new policy choice.
6. Continue through documentation or route authorized exact-target metadata repair to `pui-collaborate` when scope, live permission, idempotency, and rollback are available. Isolate a conflicting item without stopping unrelated portfolio work.
7. Return a dated evidence matrix, gaps, uncertainty, and one explicit next transition. Request a decision only when policy or product direction is genuinely unresolved, or when the action changes access, secrets, rulesets, security disclosure, publication, release, or another privileged or difficult-to-reverse surface.

Remain read-only in this leaf and hand authorized mutations to their registered leaves. Missing Project or administration scope is an observation limit that triggers narrower evidence or a scoped no-action result, not proof that configuration is absent.

Return one handoff conforming to `internal/agent-operations/schemas/skill-handoff.schema.json`, with `fromId` set to `pui-govern`, the registered governance report artifact, and at most one `nextSkillId`.

Communicate with the user in the user's current language. Keep GitHub names and identifiers canonical.

Files in this skill

  • SKILL.md2.3 KB
  • agents/openai.yaml277 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.