factory-watch
Read the factory's status once and report a run that stopped, a PR that opened, or a failing precondition. Use when you type /factory-watch for one read. Manual only, so a /loop fire cannot run it: to watch on a loop, type the plain prompt `/loop 30m run bin/factory status (bin/runner 'bin/factory status' off the runner) and report only a run that stopped, a new pr-opened, or a new precondition FAIL since the last check`. Not for starting, stopping or merging a run: use bin/factory itself.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- factory-watch
- description
- Read the factory's status once and report a run that stopped, a PR that opened, or a failing precondition. Use when you type /factory-watch for one read. Manual only, so a /loop fire cannot run it: to watch on a loop, type the plain prompt `/loop 30m run bin/factory status (bin/runner 'bin/factory status' off the runner) and report only a run that stopped, a new pr-opened, or a new precondition FAIL since the last check`. Not for starting, stopping or merging a run: use bin/factory itself.
- argument-hint
- [-C <repo> when the factory targets another repo]
- disable-model-invocation
- true
03 SKILL.md 正文
factory-watch
Trigger: user types /factory-watch. Manual-only (disable-model-invocation: true).
It reads; it never starts, stops or merges a run, and never runs bin/factory run, stop, merge or next.
- Run
bin/factory statusonce, as one bare command from the factory's checkout on the runner, where the runs and the exports live. From any other host runbin/runner 'bin/factory status'; otherwise the table is empty and the preconditions fail falsely. When$ARGUMENTSholds-C <repo>, runbin/factory -C <repo> statusinstead; off the runner,bin/runner 'bin/factory -C <repo> status'with the runner's path for<repo>. - Compare its output with the previous call in this session.
A run has stopped when its status is anything but
running,waiting-limitorpr-opened:stopped*,stuck,no-change,ticket-defect,abandonor another exit. On the first call there is none: print one line counting the runs by status (as in step 3), each stopped run, and eachFAILline under preconditions, and keep this output as the baseline. Something is new when a run stops, a run reachespr-opened, or aFAILline appears under preconditions. - Print one line per new thing, leading with the ticket and what the owner should do, and one line counting the runs by status. Print nothing else.
If the command fails, say so in one line with its exit code and stop; the next call tries again.
04 同仓库更多 Skills
verify-loam
Prove a Loam change in a real rendered project: copier-render this checkout, then drive headless and interactive Claude Code sessions, stubbed Codex runs and copier updates in the render and capture hook, memory, deny-rule, update and check evidence. Use before shipping any seed, hook, settings or copier.yml change.
hypothesis-tree
Persistent hypothesis tree for multi-step, multi-session investigations. Use when a "why does X fail" question branches into sub-hypotheses that must survive across sessions, or when falsifiable claims need tracking with dated, linked evidence. Maintains one markdown file with add, update, review, and prune subcommands, each gated on verifying the cited evidence file exists and supports the claim. NOT for a single-session bug hunt, a one-command question, or general note-taking.
fable-prompting
Which Claude Fable 5.2 guide sections you can act on from a session, and which Claude Code already injects. Use before writing a Fable session prompt or brief.
catchup
Read-only session bootstrap briefing of git state, commits, handoff, tasks and red flags. Use when resuming work, starting a fresh session, or unsure of project state.
writing-for-agents
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
wizard
Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.