
from gsd-skill-creator51
Autonomously triage GitHub issues, submit fixes as PRs, and perform adversarial PR reviews to find security and quality problems.
This workflow-driven skill scans open GitHub issues, triages them to a terminal state (duplicate, awaiting info, PR submitted, or escalated), and where appropriate generates fixes and submits PRs. After fixes complete it runs an adversarial review across open PRs to detect security, supply-chain, and logic issues.
Trigger when you want automated, policy-driven maintenance of a repo: mass-triage backlog, generate reproducible test-first fixes, enforce PR discipline, or run security-minded code reviews across many PRs. Not for cosmetic edits or tasks requiring human-only judgement.
Best used with agent runtimes that support code changes and GitHub operations (Claude Code, Codex/Code models, GitHub gh CLI-enabled agents).
Issue triage and PR review workflow skill with strong security posture — explicit prompt injection guards, spam detection with flag-only policy, and worktree isolation. Well-structured SKILL.md with clear terminal states and test-first discipline for bugs. No bundled scripts; relies entirely on gh CLI and agent orchestration. Shell snippets in monitoring section lack error handling. Solid workflow skill for repos needing automated triage.
No scripts to execute — static analysis only. Prompt injection guard and spam detection sections are noteworthy security-conscious design choices. Communication policy (never comment on effort/scope) is a practical touch.
Security Hygiene
Security hygiene guidelines for self-modifying agent systems: path sanitization, safe YAML handling, data poisoning checks, and staging/quarantine practices to
Waves — Periodic Phenomena
Conceptual skill covering wave mechanics and frequency analysis (SHM, wave equation, Fourier concepts) for agent reasoning and math tooling.