Explore skills

Run comprehensive code quality checks with ruff (format, lint) and ty (type checking). Use when checking code quality, fixing linting errors, or ensuring code follows best practices before commits or PRs.

Profile ML training performance to identify bottlenecks (data loading, compute, memory usage) and optimize GPU utilization. Use when training is slow, GPU utilization is low, or experiencing memory issues.

Run adversarial audit pipeline against shipped code

Query, browse, and manage the project capability index

Review codebase quality — find stale decisions, knowledge gaps, and implicit dependencies

Single entry point for all architecture decision operations

Interactive walkthrough to clean up stale feature directories

Archive feature working files after a PR has merged

Parallel batch coordinator for independent work units

Analyse the feature brief to identify relevant KB topics and ADRs

Adversarial test hardening — domain-lens behavioral attacks on contracts before implementation

Implement stubs until all tests pass