Claude Skills by baekenough

Multi-agent structured debate with anti-groupthink mechanisms — Devil's Advocate, minority opinion protection, 2-round hard cap. Use when divergent perspectives matter more than consensus (architecture decisions, design tradeoffs, contested specifications).

Execute CLI commands through RTK proxy for token-optimized output

Idiomatic Rust patterns from official guidelines

Full R017 verification (5+3 rounds) before commit

Analyze external URL to evaluate fit with oh-my-customcode project and auto-create GitHub issue with verdict

Spec-Driven Development workflow — enforces sdd/ folder hierarchy with planning-first gates, current-state artifacts, and completion verification

Alias for sdd-dev — Spec-Driven Development workflow

Pre-flight security analysis before code changes — identifies trust boundary risks, dangerous patterns, and provides proceed/caution/block advisory before implementation starts

Routes agent management tasks to the correct manager agent. Use when user requests agent creation, updates, audits, git operations, or verification.

Semble (MinishLab) MCP wrapper — semantic code search via embeddings (98% token reduction vs grep+read, NDCG@10=0.854)

Analyze task trajectories to propose reusable SKILL.md candidates from successful patterns

Search and install skills from skills.sh marketplace when internal skills are insufficient

Snowflake best practices for cloud data warehouse design, query optimization, and cost management

Apache Spark 4.0.2 best practices for PySpark and Scala distributed data processing

Spring Boot 4.0 patterns for enterprise Java 25 applications

Show system status and health checks

6-stage structured development cycle with stage-based tool restrictions

Detect stuck loops and advise recovery strategies

PostgreSQL performance optimization guidelines from Supabase. Apply when writing SQL, designing schemas, configuring RLS, or optimizing database performance.

Use when encountering any bug, test failure, or unexpected behavior. Enforces a strict reproduce-first, root-cause-first, failing-test-first debugging workflow before fixing.

Auto-decompose large tasks into DAG-compatible parallel subtasks

Three-layer token defense stack — audit current settings, apply safe/CI levers, and monitor status

Type-safe TypeScript patterns from industry standards

Sync documentation with project structure

Update agents from external sources (GitHub, docs, etc.)

Deploy applications to Vercel with auto-detection and preview URLs

UI code review with 100+ rules for accessibility, performance, and UX

Use the project wiki as RAG knowledge source — search wiki pages to answer codebase questions before exploring raw files

Generate and maintain a persistent codebase wiki — LLM-built interlinked markdown knowledge base (Karpathy LLM Wiki pattern)

Worker-Reviewer iterative pipeline for quality-critical code with review cycles

A mock core skill for testing

A mock harness skill for testing

A mock package skill for testing