agentleFS
fmindfmind/dot/skills/upgrade-tools/SKILL.md

upgrade-tools

Skill13 starsChanged yesterdayMITView on GitHub

Upgrade one repo's mise tools, uv deps, Actions, and images to latest stable.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • 1

    Very specific

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

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

    Keep lockfiles authoritative: commit mise.lock, uv.lock, and .terraform.lock.hcl, and pin Actions by 40-hex SHA and images by @sha256: …

    line 34
  2. Distinctive:

    Document held pins: keep inline comments explaining every below-latest hold (ABI limits, broken upstream assets) and re-verify on each pass.

    line 36

≈ “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

Upgrade Tools

Upgrade a single repository's tools and dependencies to latest stable releases in dependency order, gating each layer (playbook owns commands; mise owns pin rules).

Workflow

  1. Scope to one repository (default .): never bump another repo unless the user gives an explicit list, canonicalizing paths/worktrees and qualifying each independently.
  2. Separate read-only audits from upgrades: disk/drift audits inspect declarations, locks, backends, and installed versions (including .git-less restored folders) without resolving or installing anything.
  3. Require a green baseline: pass mise run check and mise run test before bumping (isolate dirty trees via git-worktree; stop and report if baseline is red).
  4. Upgrade in layer order, stopping to diagnose on any failure and restoring prior pins if unqualified:
    • mise: mise run upgrade (or mise run lock:tool TOOL...) in fmind/dot; exact stable pins per tool-versions.md elsewhere.
    • Python: pyproject.toml and uv.lock per playbook.md.
    • Infra & images: OpenTofu providers and OCI base/fixture digests.
    • CI & formatters: GitHub Actions SHAs and dprint plugins.
  5. Verify the final candidate: pass the full gate and runtime smoke checks (lefthook run pre-commit --all-files only in an isolated snapshot if the source tree is dirty).
  6. Commit per ecosystem on request: chore(deps): upgrade <ecosystem> to latest with its lockfile (split mise run upgrade commits by lockfile) per commit messages.
  7. Report and preview cleanup: list old → new versions, held exceptions, gate results, and dry-run prune previews (playbook.md); deleting installs requires separate authority.

Gotchas

  • Stable releases only: skip RCs, betas, and pre-releases; keep pyproject.toml ranges distinct from exact mise.toml pins.
  • Keep lockfiles authoritative: commit mise.lock, uv.lock, and .terraform.lock.hcl, and pin Actions by 40-hex SHA and images by @sha256: digest.
  • Isolate breaking majors: uv lock --upgrade can cross majors within open constraints; review lock diffs and split breaking migrations.
  • Document held pins: keep inline comments explaining every below-latest hold (ABI limits, broken upstream assets) and re-verify on each pass.
  • Check release notes for behavior shifts: e.g. betterleaks 1.10 ignoring .betterleaks.toml without --config or --unsafe-auto-load-config (betterleaks).
  • Check upstream READMEs for in-place successors: isArchived and mise registry miss unarchived handoffs (actionlint → jactionlint, gitleaks → betterleaks).

Documentation

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.