メニュー
Use when handing work to another peer, deciding whether to spawn or reuse a peer, briefing a lane, asking for a plan before code, or using a second peer for critique or review.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/prassanna-ravishankar/repowire --skill delegationコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Integration test for repowire peer-to-peer messaging. Supports claude-code, opencode, or mixed-agent-type testing with circle boundaries and cross-agent-type communication. Can run all modes in parallel via agent teams.
Parallel divergent ideation via mesh fan-out. Spawns N frame-shifted peers, scores results, deepens top survivors. Use on open-ended design choices, fuzzy bugs, naming, API surfaces, or explicit /adhd invocation. Skip for syntax, lookups, or closed prompts ("quick", "standard", "canonical").
Use when cleaning up peers, sessions, tmux panes, processes, feature worktrees, branches, generated artifacts, or after a delegated lane or PR is complete.
SOC 職業分類に基づく
SKILL.md を表示中
| name | delegation |
| description | Use when handing work to another peer, deciding whether to spawn or reuse a peer, briefing a lane, asking for a plan before code, or using a second peer for critique or review. |
Use this when giving work to a peer.
Reuse a peer when the work continues its existing context in the same worktree. Spawn fresh for independent review, decoupled concerns, cross-model critique, or when the current peer's context is too long.
One worktree per concern. Do not put two implementation peers on overlapping files in the same worktree.
Calibrate brief depth to the cost of getting it wrong.
Always include what success looks like and how the peer should report back.
Ask for a plan before code when a wrong approach would cost more than about 30 minutes to unwind. For architectural-but-bounded plans, use a different-model peer to critique before implementation.
Critique brief: include the proposed plan, files to inspect, risks to probe, and ask for skeptical-but-fair feedback with citations. Relay the critique back to the implementing peer and ask them to absorb or push back with reasoning.
Use a different peer for review when stakes warrant. CI green is not feature verification; smoke the actual behavior for UI, API, or workflow changes.
