菜单
Use when the user asks for durable or recurring background work, work that should survive peer death, retry/cancel/result tracking, spawned-on-demand workers, agent folders, daily briefs, or deciding between repowire jobs and schedule.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
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").
Get an independent implementation plan from a DIFFERENT AI agent over the repowire mesh before you build (e.g. have Gemini or Codex draft an approach for Claude to critique). Use when you want a second perspective on how to approach a task.
Get a code/PR/plan review from a DIFFERENT AI agent over the repowire mesh (e.g. have Codex review Claude's work). Use when you want an independent second opinion from another backend before merging or committing.
Hand a task to another AI agent peer over the repowire mesh — reuse an existing peer on the chosen backend, or spawn one if none exists. Use when you want to offload work to a specific agent (e.g. delegate a refactor to a Codex peer) and track it to completion.
Install or update repowire and its skill pack from inside an agent session — ensure the repowire CLI/hooks are set up, install the skills, and verify the mesh works. Use on a fresh machine or to refresh skills to the latest.
Reference for how to use the repowire mesh — ask/ack vs notify, broadcast, peer discovery, spawning, and cross-agent workflows. Use when you need to coordinate with other AI agents over repowire and want the right primitive for the job.
| name | durable-jobs |
| description | Use when the user asks for durable or recurring background work, work that should survive peer death, retry/cancel/result tracking, spawned-on-demand workers, agent folders, daily briefs, or deciding between repowire jobs and schedule. |
Use repowire jobs when work needs durable lifecycle state, recovery after peer death, retry/cancel/result inspection, spawned-on-demand execution, or recurrence. Use schedule only for future ask/notify wakeups to an existing target.
Agent folders are a convention, not a registry. Scaffold them with:
repowire agents create <name> --backend <runtime>
Then create jobs targeting the folder's absolute path with --path <abs-path> --backend <runtime>. The folder's AGENTS.md is the source of truth; CLAUDE.md is only a shim for Claude Code. Other supported runtimes load AGENTS.md directly.
--result-surface is metadata only until delivery routing exists. Do not claim jobs send Telegram, email, or dashboard notifications automatically. Workers must update job results explicitly.
User intent: "Every morning, summarize important email and send me a brief."
Route:
repowire agents create daily-email-brief --backend codex
repowire jobs create "Daily email brief" \
--path "$(pwd)/.repowire/agents/daily-email-brief" \
--backend codex \
--cron "0 8 * * *" \
--prompt "Prepare today's email brief. Use the job_id and attempt_id from this prompt when updating lifecycle state." \
--result-surface telegram
Put standing worker guidance in .repowire/agents/daily-email-brief/AGENTS.md: email tool expectations, privacy boundaries, what counts as important, and output format. Keep credentials outside the folder and outside job records.
repowire jobs create "Inspect billing webhook compatibility" \
--path /path/to/billing-repo \
--backend codex \
--prompt "Inspect webhook compatibility risk and update this job with a concise result."
repowire jobs run <job_id>
Use a job instead of notify_peer when the result must survive peer death, be retried, or be inspected later with repowire jobs result.
repowire schedule self 30m "Check whether the release peer replied."
Use schedule, not jobs, when this is just a future message to an existing session.