agentleFS
jimtinjimtin/production-ai/skills/laptop-currency-maintenance/SKILL.md

laptop-currency-maintenance

Skill1 starsChanged 5 months agoMITView on GitHub

Keep this macOS developer laptop current without unsafe repo mutation. Use when asked to update Homebrew, PowerShell, Vercel CLI, Node, gh, dotnet, global developer tooling, package currency, stale CLIs, or to keep the laptop up to date while safely auditing local repo dependencies.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 2 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start with a read-only audit before any update.

    line 14
  2. Very specific:

    Use the bundled automation tool: scripts/laptop-currency-maintenance.mjs (installed at …

    line 15
  3. Very specific:

    Do not use sudo.

    line 17
  4. Distinctive:

    Do not auto-update Homebrew casks, macOS system updates, App Store apps, Docker Desktop, repo dependencies, lockfiles, language runtime …

    line 18
  5. Very specific:

    Use completed_with_warnings when audits complete but checks are unavailable or fail.

    line 21
  6. Very specific:

    Treat repo dependency upgrades as implementation work: they need repo-local planning, $test-readiness-preflight, local/container …

    line 22
  7. Very specific:

    Audit first: Run laptop-currency-maintenance.mjs audit or update --dry-run to collect Homebrew, global npm, repo package, audit, CLI …

    line 28
  8. Very specific:

    Apply safe host updates only: If updating, run the tool in update mode.

    line 30
  9. Very specific:

    Escalate repo upgrades separately: If the user asks to upgrade repo dependencies, switch to repo-specific work and use …

    line 32

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Laptop Currency Maintenance

Purpose

Use this skill to keep host developer tooling current while preserving repo safety. The default v1 boundary is strict: Homebrew formulae may be updated by the approved automation, while repo package manifests and lockfiles are audited only unless the user starts a separate repo-specific upgrade task.

Operating Rules

  • Start with a read-only audit before any update.
  • Use the bundled automation tool: scripts/laptop-currency-maintenance.mjs (installed at ~/.codex/skills/laptop-currency-maintenance/scripts/). Copy scripts/config.example.json to config.json beside it and set absolute paths — ~ is not expanded.
  • Auto-update only unpinned Homebrew formulae in v1.
  • Do not use sudo.
  • Do not auto-update Homebrew casks, macOS system updates, App Store apps, Docker Desktop, repo dependencies, lockfiles, language runtime managers, or global npm packages.
  • Skip and report pinned Homebrew formulae, unavailable tools, casks, and failed checks.
  • Highlight configured highImpactFormulae in reports when they are outdated.
  • Use completed_with_warnings when audits complete but checks are unavailable or fail.
  • Treat repo dependency upgrades as implementation work: they need repo-local planning, $test-readiness-preflight, local/container validation, dependency audit, repo-scoped containerized gitleaks, and $security-threat-model when push/readiness is in scope.
  • Disk cleanup is not package currency proof. Use this skill for version freshness and package drift.
  • Sanitize reports and Discord messages. Do not print tokens, auth headers, env values, provider IDs, or secret-looking strings.

Workflow

  1. Audit first. Run laptop-currency-maintenance.mjs audit or update --dry-run to collect Homebrew, global npm, repo package, audit, CLI version, and cleanup evidence.
  2. Classify host tooling. Separate auto-upgradeable Homebrew formulae from pinned formulae, casks, missing tools, and manual-only update sources.
  3. Apply safe host updates only. If updating, run the tool in update mode. It refreshes Homebrew metadata, upgrades unpinned formulae, proves before/after versions, then runs brew cleanup.
  4. Audit repos without mutating them. Report outdated repo dependencies by repo, package manager, patch/minor/major gap, security audit status, dirty/active status, and recommended next action.
  5. Escalate repo upgrades separately. If the user asks to upgrade repo dependencies, switch to repo-specific work and use $test-readiness-preflight before expensive validation. Include $security-threat-model when the changed scope touches security-sensitive dependencies or push/readiness is in scope.
  6. Report completion. Include exact commands, pass/fail status, upgraded formulae, skipped items, before/after versions, report paths, Discord status, and any repo dependency proposals.

Automation Contract

The daily Laptop Currency Maintenance cron should run:

node ~/.codex/skills/laptop-currency-maintenance/scripts/laptop-currency-maintenance.mjs update

It should post to the configured Discord channel (<DISCORD_CHANNEL_ID> in the config), attach the Markdown report, and fail closed on command errors. It must not mutate repos.

Completion Blockers

  • Audit did not run before update.
  • A Homebrew command failed and the report hides or ignores the failure.
  • The tool attempts to upgrade casks, repo packages, lockfiles, global npm packages, macOS updates, App Store apps, or Docker Desktop.
  • A repo dependency recommendation is reported as completed without repo-local validation.
  • Report output includes unredacted secrets or provider identifiers.
  • Report output exposes local absolute home paths in Discord or Markdown summaries.

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.