Skip to content
Back to skills

Ztgc Improve Advisor Codebase Audit

ASecurity

Focused Perfect_AI shadcn Improve Advisor specialist: advisor codebase audit

  • 14 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 7, 2026
testinggit

Security analysis

A100/100

Scanned November 7, 2026

npx -y skills add Realjustawall/Perfect-AI --skill ztgc-improve-advisor-codebase-audit --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Ztgc Improve Advisor Codebase Audit?

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

Security grade badge for Ztgc Improve Advisor Codebase Audit
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/realjustawall-ztgc-improve-advisor-codebase-audit/badge)](https://www.skillsdirectory.com/skills/realjustawall-ztgc-improve-advisor-codebase-audit)

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: ztgc-improve-advisor-codebase-audit
description: "Focused Perfect_AI shadcn Improve Advisor specialist: advisor codebase audit"
---

# Advisor Codebase Audit

## Responsibility
Build evidence table per finding: file, line, reproduction, impact, confidence, dependencies and quick verification.

## Entry conditions
- Load the parent `$ztgc-shadcn-improve-advisor` only when shadcn Improve Advisor is relevant to this task.
- Collect concrete routes/files, user objective, dependency versions and current behavior.
- Do not rename or delete old skills or overwrite existing project content.

## Method
1. Inventory architecture, problems, test harness, dependency versions, open issues and current behavior.
2. Identify problems with line/file evidence and reproduce high-impact defects where feasible.
3. Prioritize by user value, confidence, implementation cost, regression risk and test availability.
4. Write self-contained plans to a NEW plans/ or advisor-plans/ destination without changing application code.
5. Hand off acceptance criteria, edge cases, rollback and verification commands; keep unresolved uncertainty visible.

## Specialized verification
- [ ] Never modify source files while operating as advisor
- [ ] No speculative claims without source or reproduction
- [ ] Distinct changes isolated with an acceptance checklist

## Tool usage and artifacts
- Tools: None, source-read-only advice mode.
- Evidence: baseline, exact reproduction command, before/after details, screenshot or console logs when relevant.
- Minimum outcome: separate PASS/FAIL/NOT_RUN records, with a description of any unavailable runtime or device.
- Reference: `https://github.com/shadcn/improve/blob/main/skills/improve/SKILL.md`. This locally authored guidance is not a copied GitHub Skill.

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.