agentleFS
fpindejfpindej/netrock/.claude/skills/review-dependabot/SKILL.md

review-dependabot

Skill234 starsChanged 3 months agoMITView on GitHub

Reviews a Dependabot PR and evaluates whether the dependency update is safe to merge. Use when a Dependabot PR is opened, when evaluating dependency updates, or when asked to check if a version bump is safe.

  • Installs packages

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    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:

    Verify the author is app/dependabot: If it is not, STOP and report - this skill fetches and builds the PR's code locally, and that is …

    line 19
  2. Distinctive:

    Read the Dependabot PR body for release notes / changelog links

    line 42
  3. Distinctive:

    If the PR body includes a changelog summary, check for breaking changes, deprecations, and behavioral changes

    line 43
  4. Distinctive:

    Search the codebase for direct imports/usages of the package:

    line 48
  5. Distinctive:

    Check if the package is a transitive dependency (only in lock file, not directly referenced)

    line 51
  6. Very specific:

    Never gh pr checkout - that would switch the user's working tree.

    line 56
  7. Distinctive:

    If tests pass, that's a strong signal.

    line 67
  8. Distinctive:

    Always run the test suite - never skip it

    line 99
  9. Distinctive:

    When in doubt, verdict is NEEDS MANUAL REVIEW, not SAFE TO MERGE

    line 100

≈ “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

Reviews a Dependabot PR and evaluates whether it is safe to merge.

Argument: PR number or URL.

Open Dependabot PRs: !gh pr list --author 'app/dependabot' --json number,title --jq '.[] | "#\(.number) \(.title)"' 2>/dev/null | head -10 || echo "(none found)"

Steps

  1. Fetch PR metadata: gh pr view {number} --json number,title,headRefName,body,labels,author
  2. Verify the author is app/dependabot. If it is not, STOP and report - this skill fetches and builds the PR's code locally, and that is only acceptable for Dependabot-authored version bumps. Never proceed for an arbitrary PR number.
  3. Get the diff: gh pr diff {number}
  4. Identify the package, old version, new version, and version bump type (patch / minor / major)

Determine scope

SignalHow to check
Backend (.NET)Diff touches Directory.Packages.props
Frontend (pnpm)Diff touches package.json / pnpm-lock.yaml
Dev-only dependencyPackage appears only in test projects, build tooling, or devDependencies
Production dependencyPackage is imported/used in shipped code

Evaluate risk

1. Version bump type

  • Patch (x.y.Z) - bug fixes, low risk
  • Minor (x.Y.0) - new features, backwards-compatible by semver, low-medium risk
  • Major (X.0.0) - potentially breaking, high risk - requires careful review

2. Changelog review

  • Read the Dependabot PR body for release notes / changelog links
  • If the PR body includes a changelog summary, check for breaking changes, deprecations, and behavioral changes
  • For major bumps: WebFetch the changelog URL if provided

3. Usage impact

  • Search the codebase for direct imports/usages of the package:
    • .NET: grep -r "using.*{PackageNamespace}" src/backend/ and check .csproj references
    • Frontend: grep -r "from ['\"].*{package}" src/frontend/src/
  • Check if the package is a transitive dependency (only in lock file, not directly referenced)
  • For major bumps: check if any API surface we use was deprecated or removed

4. Test verification

Never gh pr checkout - that would switch the user's working tree. Test in a disposable worktree:

git fetch origin pull/{number}/head
git worktree add /tmp/dependabot-{number} FETCH_HEAD
  • Backend: dotnet build /tmp/dependabot-{number}/src/backend/MyProject.slnx && dotnet test /tmp/dependabot-{number}/src/backend/MyProject.slnx -c Release
  • Frontend: cd /tmp/dependabot-{number}/src/frontend && pnpm install --ignore-scripts && pnpm run test && pnpm run check
    • --ignore-scripts is deliberate: package lifecycle scripts are the main local code-execution risk of building PR content. If the updated package genuinely needs its install scripts to function, do NOT run them - verdict is NEEDS MANUAL REVIEW.
  • Afterwards: git worktree remove --force /tmp/dependabot-{number}
  • If tests pass, that's a strong signal. If they fail, identify whether the failure is related to the update.

Output format

## Dependabot Review: {package} {old} -> {new}

**Bump type**: patch / minor / major
**Scope**: backend / frontend
**Dependency type**: production / dev-only
**Test result**: PASS / FAIL

### Risk assessment

{Summary of findings - changelog highlights, breaking changes, usage impact}

### Verdict: SAFE TO MERGE / NEEDS MANUAL REVIEW / DO NOT MERGE

{Reasoning for verdict}

Verdict criteria

VerdictWhen
SAFE TO MERGEPatch/minor bump, tests pass, no breaking changes in changelog, no security advisories
NEEDS MANUAL REVIEWMajor bump with passing tests, or minor bump with deprecation warnings, or unfamiliar package with wide usage
DO NOT MERGETests fail due to the update, known breaking changes that affect our usage, security concerns in the new version

Rules

  • Research only - never modify the user's working tree; file side effects stay inside the disposable worktree. Building/testing still EXECUTES code from the PR (MSBuild targets, test code) with your privileges - that is why the author check and --ignore-scripts are mandatory, and why this skill is for Dependabot PRs only
  • Always run the test suite - never skip it
  • When in doubt, verdict is NEEDS MANUAL REVIEW, not SAFE TO MERGE
  • A passing test suite does not override known breaking changes in the changelog

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.