agentleFS
Sign inStart free
d-o-hubd-o-hub/github-template-ai-agents/.agents/skills/dependency-upgrades/SKILL.md

dependency-upgrades

Skill2 starsChanged 10 days agoMITView on GitHub

Perform routine, low-risk dependency version bumps and vulnerability-driven upgrades (patch/minor, or major with no source changes). Use when updating packages, action pins, pre-commit hooks, lockfiles, or base images — even if they just say "bump lodash" or "fix this CVE". Not for migration-refactoring (major bumps that force source changes, framework or language migrations).

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 3

    Distinctive

    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
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Never bump major versions together with unrelated changes.

    line 35
  2. Distinctive:

    Never ignore a lockfile diff — read it.

    line 36
  3. Distinctive:

    Never merge a security upgrade red (failing checks void the point).

    line 38

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “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

Dependency Upgrades

Routine bumps and CVE-driven upgrades: one dependency per change, verified by CI, with a rollback path.

When to Use

  • Bumping a package, action version, or base image
  • Vulnerability alert (CVE, Dependabot, audit finding)
  • Lockfile refresh or stale pins

Steps

  1. Scope one upgrade — one dependency per PR/branch so failures attribute cleanly. Group only provably-coupled pairs.
  2. Read the changelog — note breaking changes and migration notes before bumping, not after CI fails.
  3. Bump + lock — update the manifest and regenerate the lockfile with the project's canonical tool; never hand-edit lockfiles.
  4. Verify — full relevant test suite plus the quality gate; for security bumps, confirm the CVE is actually remediated (audit clean).
  5. Record rollback — pin the previous version in the PR body so revert is one commit.

Rules

  • Never bump major versions together with unrelated changes.
  • Never ignore a lockfile diff — read it; unexpected transitive churn is a signal, not noise.
  • Never merge a security upgrade red (failing checks void the point).
  • This repo's Dependabot groups all action minor/patch bumps into one PR (.github/dependabot.yml) and auto-merges via ADR-032 when MAINTAINER_AUTOMATION=true; never let that bypass the verify step.

Rationalizations

RationalizationReality
"It's just a patch bump, skip the tests"Patch releases introduce regressions regularly; the suite is cheaper than the incident.
"Dependabot already tested it"Bot CI covers the bot's matrix, not your integration paths; run your gates.

Red Flags

  • Major bump bundled with feature work
  • Hand-edited lockfile
  • CVE upgrade merged with failing checks

See Also

  • migration-refactoring — Framework and language migrations
  • test-runner — Execute tests after the bump
  • git-github-workflow — Ship the upgrade end-to-end

References

  • evals/evals.json — upgrade cases incl. breaking-change negative

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.

Check your setup, free