skills for everythingdeep-think
IF complex problem needing multi-perspective analysis THEN use this. 5 agents, 3 phases: understand→synthesize→recommend. No implementation. For competing implementations → tournament.
SKILL.md
| Name | deep-think |
| Description | IF complex problem needing multi-perspective analysis THEN use this. 5 agents, 3 phases: understand→synthesize→recommend. No implementation. For competing implementations → tournament. |
ancplua-claude-plugins
Claude Code plugin marketplace. 8 plugins for parallel agent orchestration, quality gates, and automated enforcement in Claude Code sessions.
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
| 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 |
| otelwiki | Built-in reference for OpenTelemetry monitoring standards so the AI writes correct telemetry code instead of guessing |
| 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 |
| ancplua-project-routing | Automatically recognizes what kind of project you're in and loads the right tools and rules. No configuration needed |
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
Technical details
8 plugins, 23 commands, 4 skills, 14 agents, 21 scripts, 7 hook configs.
Tri-AI review system: Claude, Copilot, CodeRabbit all review PRs independently.
plugins/
├── exodia/ # parallel agent orchestration (9 commands + 2 skills)
├── metacognitive-guard/ # quality gates + commit integrity + CI verification
├── otelwiki/ # OpenTelemetry docs + sync
├── hookify/ # user-configurable behavior rules
├── feature-dev/ # guided feature development + code review
├── council/ # five-agent council (Opus captain + Sonnet specialists + Haiku janitor)
├── dotnet-architecture-lint/# .NET build pattern enforcement
└── ancplua-project-routing/ # project-aware agent routing