agy-jobs

1<!-- Generated from skills/jobs/SKILL.md; run npm run generate:opencode. Do not edit here. -->
2 
3# agy jobs
4 
5Manage background staffer/research/review/implement jobs. State is per repository in `.agy-staff/` (git-excluded via `.git/info/exclude`). Always invoke the companion from the target repository so AGY attaches that repository as its workspace; `.agy-staff/` is companion state rather than a repository source modification, so do not relocate companion calls to `/tmp` or delete `.agy-staff/`. Only ask runs synchronously.
6 
7This file lives at `<plugin-root>/opencode-skills/agy-jobs/SKILL.md`:
8 
9```bash
10node "<skill-dir>/../../companion/agy-companion.mjs" <command> [args]
11```
12 
13## Collect the result
14 
15Default flow: prepare the prompt → dispatch → wait for the final result → validate as needed. While a job is running, do not proactively call `observe`/`status`, read logs or inspect intermediate artifacts. Do not query progress for routine updates or create sleep/observe loops. Observe only when the user explicitly asks for progress; diagnose after receiving a failure or a result requiring intervention.
16 
171. Keep the returned job id. Run `wait <id> --timeout 10m` using the same unsandboxed context as launch (in the background when the host supports background sessions or completion delivery, or directly in the foreground when the host only provides a synchronous shell tool). Use a separate wait for each job; never wait for several jobs serially in one shell.
18…