skills for everythingred-blue-review
Use for adversarial security/quality review before release. Red team attacks, Blue team defends. For broad audit -> use mega-swarm.
SKILL.md
| Name | red-blue-review |
| Description | Use for adversarial security/quality review before release. Red team attacks, Blue team defends. For broad audit -> use mega-swarm. |
ancplua-claude-plugins
Claude Code plugin marketplace. 19 plugins for parallel agent orchestration, quality gates, automated enforcement, OpenTelemetry semconv injection, .NET release driving, project-memory curation, and session debriefs.
What this does
Spawns up to 12 parallel subagents for audits, fixes, and reviews. Each phase is gated — work only advances when the gate passes. No manual babysitting.
Plugins
Sample of available plugins (full list: 19 plugins, 31 commands, 20 skills, 26 agents):
| Plugin | What it does in plain language |
|---|---|
| exodia | Summons up to 12 AI workers that tackle different parts of your code simultaneously. One finds problems, another fixes them, another reviews the fix — all at the same time |
| metacognitive-guard | Watches the AI while it works. If it's about to cut corners, guess instead of verify, or say "done" when it isn't — this blocks it before the mistake happens |
| feature-dev | A guided process for building new things: understand what exists, design the plan, build it, review it. No skipping steps |
| hookify | Custom tripwires you set up. "Never do X" or "Always check Y before Z." The AI physically cannot proceed if it violates your rules |
| dotnet-architecture-lint | Enforces .NET project structure rules automatically — catches version mismatches and structural violations before they ship |
| council | Five-agent council for complex tasks. Opus captain decomposes and dispatches, three Sonnet specialists research/synthesize/check, Haiku janitor flags bloat |
| design-studio | Design intelligence helpers for visual direction, UX patterns, and design-system generation |
| elegance-pipeline | Multi-agent pipeline that scores code elegance and optionally refactors the weakest files through gated stages |
| code-simplifier | Simplifies implementations while preserving behavior — measures elegance as problem-complexity / solution-complexity |
| claudemd-curator | Audits and curates project-memory artifacts such as CLAUDE.md, AGENTS.md, and .claude/rules/*.md |
| session-debrief | Generates self-contained HTML reports from Claude Code session transcripts: tokens, cache, subagents, skills, and expensive prompts |
How does this work without failing?
Every step is a gate. Work only moves forward if the gate says PROCEED. If it says HALT — the work stops, gets diagnosed, and gets fixed before anything else continues. No hoping. No skipping. No "it probably works."
Install
Add the marketplace, then install plugins individually:
# Add the marketplace
/plugin marketplace add ANcpLua/ancplua-claude-plugins
# Install plugins you want
/plugin install exodia@ancplua-claude-plugins
/plugin install metacognitive-guard@ancplua-claude-plugins
/plugin install hookify@ancplua-claude-plugins
/plugin install claudemd-curator@ancplua-claude-plugins
/plugin install session-debrief@ancplua-claude-plugins
Technical details
19 plugins, 31 commands, 20 skills, 26 agents.
Tri-AI review system: Claude, Copilot, and CodeRabbit review PRs independently.
GitHub review automation
Codex review lives in .github/workflows/codex-code-review.yml.
- Set the
OPENAI_API_KEYrepository secret to enable the workflow - Codex runs in a
read-onlysandbox withdrop-sudo - The workflow posts a formal PR review from structured Codex output
- Self-review is blocked when a PR only changes Codex review automation files
plugins/
├── exodia/ # parallel agent orchestration (9 commands + 2 skills)
├── metacognitive-guard/ # quality gates + commit integrity + CI verification
├── hookify/ # user-configurable behavior rules
├── feature-dev/ # guided feature development + code review
├── council/ # five-agent council (Teams API)
├── dotnet-architecture-lint/# .NET build pattern enforcement
├── design-studio/ # design intelligence + BM25 search
├── elegance-pipeline/ # code elegance scoring + gated refactoring
├── code-simplifier/ # complexity reduction agent
├── claudemd-curator/ # project-memory curation
├── session-debrief/ # Claude Code session usage reports
└── otelhook/ # OTel GenAI semconv passive context