Multi-model consensus system — 3 LLMs cross-examine each other to catch blind spots on critical decisions
npx claudepluginhub supportersimulator/3-surgeons --plugin 3-surgeonsChallenge a proposed action with the neurologist corrigibility skeptic — tests counter-positions before committing
Run a full 3-phase cross-examination on a topic — independent analysis, cross-review, synthesis
Set up the 4-folder document system (inbox/vision/reflect/dao) in this project
Run the gains gate verification — infrastructure health checks that must pass before proceeding
Health check all three surgeons to verify they are reachable and operational
Scan content for complexity vectors — detects authentication, security, performance, concurrency, and other risk dimensions
Invoke the 3-Surgeons system for multi-model consensus
Verify all three surgeons are reachable and operational
Enforce system health and safety invariants via GainsGate, CardioGate, and CorrigibilityGate
Full A/B test lifecycle with safety constraints and human veto windows
Rapid 3-surgeon verdict on whether a proposed fix or change is sound
HARD-GATE — multi-model review of architectural decisions before implementation proceeds
External-model cross-examination review with optional git context for code and architecture decisions
Confidence-weighted vote from multiple surgeons to validate claims and check assumptions
HARD-GATE — force adversarial counter-arguments to ensure both sides of a decision are examined
Deep 3-phase multi-model analysis — cross-examine a claim using multiple AI models
Direct answer from a specific surgeon without consensus overhead
Set up the 4-folder document system (inbox/vision/reflect/dao) in a project
Structural integrity verification — monotonic counters, service health, and evidence store operability
Metacognition — evaluate gate system effectiveness, calibrate thresholds, and fine-tune invariance
Narrate all 3-surgeons operations with surgeon visual identities — phased cross-exams, multi-surgeon consults, and single-surgeon queries
Local-model health pulse, corrigibility challenges, and surgeon capability introspection
HARD-GATE — multi-dimensional verification that implementation meets requirements before completion
HARD-GATE — adversarial review of implementation plans by all 3 surgeons before coding begins
Self-directed research on a topic — finds information, synthesizes findings, and tracks costs
Selects review depth mode for 3-surgeon cross-exam based on task risk
Detect complexity risks across configurable dimensions
Interactive API key resolution — guides users through fixing missing surgeon API keys with detected options
First-run onboarding — detect backends, configure API keys, and verify surgeon connectivity
Bootstrap skill — teaches the 3-Surgeons multi-model consensus philosophy, surgeon roles, tool access, and when to invoke other skills
Team-oriented workflow plugin with role agents, 27 specialist agents, ECC-inspired commands, layered rules, and hooks skeleton.
Admin access level
Server config contains admin-level keywords
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, rules, and legacy command shims evolved over 10+ months of intensive daily use
Semantic search for Claude Code conversations. Remember past discussions, decisions, and patterns.
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
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Uses power tools
Uses Bash, Write, or Edit tools
No model invocation
Executes directly as bash, bypassing the AI model
No model invocation
Executes directly as bash, bypassing the AI model