Use when a user wants to define done-criteria or a stop condition for an autonomous multi-turn agent run — setting a Claude Code /goal, a Definition of Done for a Codex or external-orchestrator loop, or turning a vague task into a verifiable completion…
Periodically audit a repository's docs for staleness against the implementation: run a cheap mechanical scan (dead source-path references, dead links, leftover DEPRECATED/TBD banners, doc-vs-code freshness gap, git-invisible ignored leftovers,…
Optimize CLAUDE.md files using progressive disclosure: reduce bloat, move details to references, extract reusable patterns into skills, improve context efficiency. Use when CLAUDE.md is too long or needs restructuring.
Interface design skill for dashboards, admin panels, apps, and tools. NOT for marketing design. Provides design intent, domain exploration, craft principles, and component-driven design-tool↔code sync. For PySide6/Qt technical implementation, use a…
Run PR maintenance to completion automatically after an agent creates a PR or a draft PR becomes ready for review: keep polling CI and bot review with gh, repair failures and reply in Japanese in the same worktree, escalate design loops, and squash merge when…
Maintain an agent-created pull request after creation or after draft-to-ready transition: mark reviewable draft PRs ready, poll CI and review comments with gh, repair failures in the same worktree/session, reply in Japanese, escalate design loops, and squash…
Before implementing a feature, hear out the requirement, then research existing libraries / tools / local packages (monorepo components) that already solve it. Use when scoping a new feature, evaluating build-vs-reuse, or asked to find existing solutions…
GitHub operations via gh CLI and GitHub API: create/view/merge PRs, manage issues, query API endpoints, handle workflows in enterprise or public GitHub environments.