By glubiz
Three-tier adversarial code audit system that catches AI slop, enforces code quality, and prevents people-pleasing
Adversarial code audit subagent — reviews code diffs for AI slop patterns and code quality issues with a guilty-until-proven-innocent stance
Fix regression audit subagent — reviews fix diffs to verify fixes did not introduce new slop or break existing functionality
Three-tier adversarial code review plugin for Claude Code. Catches AI slop, enforces code quality, prevents people-pleasing, and verifies CLAUDE.md compliance.
When triggered, the plugin runs three sequential tiers:
Anti-slop detection:
Code quality assurance: 10. Security (always CRITICAL) 11. Performance 12. Readability
Project rules compliance: 13. CLAUDE.md / AGENTS.md violations
Requires Claude Code and the superpowers plugin.
# Add the marketplace
/plugin marketplace add https://github.com/Glubiz/claude-code-audit.git
# Install the plugin
/plugin install claude-code-audit@claude-code-audit
# Activate
/reload-plugins
/audit
Runs the full three-tier audit against your current uncommitted changes. If no changes exist, it offers to audit the last commit.
The plugin also triggers automatically:
superpowers:code-reviewer subagent completessuperpowers:finishing-a-development-branch/audit --skip — Bypass a FAIL gate (recorded in state)/audit --last-commit — Audit the last commit when no uncommitted changes exist/audit --with-ci — Also run matching CI quality jobs via gitlab-ci-local (opt-in)The audit produces structured, machine-parseable output:
## Audit Result
**verdict:** FAIL
**blocking_issues:** 1
**total_issues:** 4
### CRITICAL
- file: src/auth.ts:42
issue: SQL injection via unsanitized user input
action: Use parameterized query instead of string concatenation
category: security
### IMPORTANT
- file: src/utils.ts:10-15
issue: formatDate helper has single caller
action: Inline into src/components/Header.ts:23
category: unjustified-abstraction
### NEXT_STEPS
1. Fix SQL injection in auth.ts
2. Inline formatDate helper
Verdicts:
PASS — No issues foundPASS_WITH_WARNINGS — IMPORTANT/MINOR issues only (advisory, does not block)FAIL — CRITICAL issues found (blocks progress until fixed)This plugin complements, not replaces, the superpowers code-reviewer. The code-reviewer focuses on correctness and architecture. This audit focuses on slop detection and quality enforcement. They run sequentially — code-reviewer first, then this audit.
MIT
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub glubiz/claude-code-audit --plugin claude-code-auditComplete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development