agentleFS
aaroncohnaaroncohn/babysitter/skills/cloud-setup/SKILL.md

cloud-setup

Skill1 starsChanged 13 days agoMITView on GitHub

Set this repo up so babysitter's skills are in the / picker of every new Conductor cloud workspace, from its first session. Either commits copies of the skills to .claude/skills/ with a daily GitHub Action that keeps them current (recommended), or keeps the plugin installed with an install script and a SessionStart hook that tells a session to /reload-plugins when plugins land late. Writes files and shows the diff; never commits. Run with `/babysitter:cloud-setup`.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. bdfinst/agentic-dev-team★ 302same name
    2. sunholo-data/ailang★ 38same name
    3. ShmalexM/agent-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository10 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read .claude/settings.json (it may not exist) and check for .claude/skills/babysitter-*/ or .claude/install-plugins.py from an earlier run.

    line 11
  2. Distinctive:

    Ask, in one message, and wait for the letter:

    line 12
  3. Very specific:

    Run python3 <this skill's dir>/cloud-setup.py apply vendored for a, or apply hook for b.

    line 15

≈ “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

cloud-setup

In a new cloud workspace, Conductor starts the first session before the repository's Setup script has installed plugins, and the app caches the / picker's list, so /babysitter: skills can be missing even after /reload-plugins. This sets the repo up so they aren't. Run it from the repo's root, in a workspace on a fresh branch.

  1. Read .claude/settings.json (it may not exist) and check for .claude/skills/babysitter-*/ or .claude/install-plugins.py from an earlier run. Say what's there.
  2. Ask, in one message, and wait for the letter:
    • a. Copies of the skills in the repo (recommended). They're in the picker from the first moment. A GitHub Action refreshes them daily and pushes any change to the default branch. They're named /babysitter-hq, /babysitter-doctor and so on (a dash, not a colon), and the plugin is turned off in this repo so nothing shows twice. Costs: a few hundred KB in the repo, and a daily commit when babysitter changes.
    • b. Keep the plugin installed. An install script for the Setup script, plus a SessionStart hook that installs what's missing and, when plugins landed after the session started, tells it to /reload-plugins. The picker may still miss them until a new chat; typing the full /babysitter:… name works.
  3. Run python3 <this skill's dir>/cloud-setup.py apply vendored for a, or apply hook for b. It writes the files and edits .claude/settings.json, keeping everything else in it.
  4. Show git status --short and the settings.json diff. Don't commit or push: tell the user to commit it on a branch and merge it the way the repo does.
  5. Then tell them what's left, which only they can do:
    • a: the workflow needs to push. If the repo or organization restricts it, allow it under Settings → Actions → General → Workflow permissions ("Read and write"), or change the workflow's last step to open a PR. If the repo requires a particular commit author, set it in that step. After merging, run it once: gh workflow run sync-babysitter-skills.yml. If the repository's Setup script in the cloud organization settings installs the babysitter plugin, remove that line.
    • b: set the repository's Setup script, in Conductor's cloud organization settings (not .conductor/settings.toml, which is local-only), to python3 .claude/install-plugins.py.
    • Either way: /babysitter:doctor (/babysitter-doctor with a) in a new cloud workspace confirms it.

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.