ワンクリックで
looper
looper には ajhcs から収集した 6 個の skills があり、リポジトリ単位の職業カバレッジとサイト内 skill 詳細ページを表示します。
このリポジトリの skills
Write compact Codex orchestration loops over accepted plans, beads, implementation slices, review/fix passes, and selective subagents. Use when the user asks for a loop, agent loop, workflow loop, Codex loop, retry loop, evaluator loop, multi-agent loop, or orchestration prompt.
Write compact Codex orchestration loops over accepted plans, beads, implementation slices, review/fix passes, and selective subagents. Use when the user asks for a loop, agent loop, workflow loop, Codex loop, retry loop, evaluator loop, multi-agent loop, or orchestration prompt.
Decompose parent beads, specs, plans, completed Wayfinder maps, and grill-with-docs output into evidence-backed, commit-sized childbeads with explicit execution packets. Use when the user asks to beadwrite, split work into atomic commits, prepare work for Looper, or turn planning artifacts into implementation units.
Inspect, create, validate, render, and deepen System Understanding Maps from typed Map Sources before planning or implementation work.
Audit open bd beads against the project's current direction and safely re-scope, defer, deduplicate, supersede, close, or delete them. Use when the user asks to clean up, prune, triage, archive, close, or delete stale, obsolete, duplicate, abandoned, or nonsensical beads.
Phone-A-SWE researches software engineering decisions before implementation. Use when the user asks to phone a SWE, choose between technical options, evaluate a library/API/framework/MCP/skill, or ground a codebase decision in repository evidence and current primary sources.