agentleFS
joaoajmatosjoaoajmatos/ze/.claude/skills/ze-pause/SKILL.md

ze-pause

Skill2 starsChanged 47 days agoUnlicenseView on GitHub

Pauses all Ze development. Every agent writes a pause handoff, then the Maestro records a snapshot and stops the agents, floors, routines, dev servers and Docker, so the machine is free for another project. Use when the user asks to pause, stop for now, or switch to another project.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never pause by dismissing agents or deleting floors: those lose the agent's connections and notes.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Pause development (Maestro only, from Ground)

The goal: nothing of Ze keeps running, and /ze-continue can pick up every thread exactly where it stopped. Run the steps in order. Do not skip the handoffs: an agent stopped without one loses its in-flight context.

  1. Snapshot. python3 scripts/process/pausectl.py snapshot writes journal/pause/<stamp>.json and .md: floors (head, ahead of main, dirty files, last review, open tasks), agents, routines, Ground dev servers with their env, and Docker services. Keep the path.
  2. Handoffs from every running agent. Send one batch (maestri ask --batch) to every agent the snapshot marks running, with a 10-minute timeout:

    PAUSE <stamp>: the user is pausing all Ze development. Within 5 minutes, stop at a safe point. Finish or abandon the current command, and start no new gates or reviews. Commit finished work on your own paths only and leave work in progress uncommitted. Append a ### <date> PAUSE <Name> [<Floor>] entry to your floor's Handoffs note ("Ground Handoffs" on Ground) with: done since your last entry, in flight (files, uncommitted or not), the exact next step, what you wait on (who or what), and commands to rerun. Update your state file journal/state/$MAESTRI_TERMINAL_ID.md the same way. Stop any dev server you started. Then reply with one line: paused: <next step>.

    Any agent that does not reply: maestri check it. If it is mid-turn after 10 minutes, press Esc (--raw '\e') and ask again. If it cannot reply (bare shell, crashed), write its entry yourself from its terminal, git status and its tasks.md rows.

  3. The Maestro's summary. Append ## Maestro summary to journal/pause/<stamp>.md with:

    • a status table with one row per floor: where it is, what the next step is, and who acts next;
    • the order in which to resume floors;
    • halted floors that must not be woken;
    • open user decisions;
    • blockers (Docker, credits, limits).

    Then update:

    • your state file;
    • the floor notes in journal/score/score.ze;
    • one line in the "Ze Roadmap" note: Paused <stamp> → journal/pause/<stamp>.md;
    • a memory project_ze_pause_<date>.md that points at the pause file (do not copy the file into it).
  4. Stop everything. python3 scripts/process/pausectl.py stop journal/pause/<stamp>.json does the following:

    • disables the enabled routines;
    • sends /exit to each running agent and records the session id and model it printed;
    • unloads every floor;
    • stops the dev servers and the leaked docker clients;
    • stops the compose services and quits Docker Desktop.

    It writes all of this back into the snapshot. Preview with --dry-run. It names any agent whose session id it could not read; that agent resumes with --continue in its own directory.

  5. Report to the user. Give the pause file path, a one-line status per floor, and anything left running (the Maestro itself, which the user closes). Tell the user to start again with /ze-continue in the Maestro terminal.

Never pause by dismissing agents or deleting floors: those lose the agent's connections and notes.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.