
Scan changed files with Semgrep for security and correctness issues. Use when verifying code before committing, reviewing implementation security, or running pre-commit quality gates.
Delegate tasks to Swival for self-reviewed code changes, sandboxed execution, secret-safe operations, cached analysis, local-model inference, and A2A agent serving or client usage. Use when a task benefits from automated review loops against acceptance criteria, filesystem sandboxing, credential encryption, LLM response caching, or orchestrating a network of A2A agents.
Structured readiness assessment before proceeding to the next environment or phase. Use when asked for a readiness review, go/no-go assessment, or deployment review.
Write and refine agent skills. Use when generating skills from docs, extracting from sessions, designing tool interfaces, managing lifecycle, or writing for dual-harness patterns.