agentleFS
kakarot-oncloudkakarot-oncloud/claude-dev-skills/skills/dependency-upgrader/SKILL.md

dependency-upgrader

Skill0 starsChanged 7 months agoMITView on GitHub

Analyzes the impact of upgrading a package or framework version — identifies breaking changes, migration steps, deprecated APIs, and risk level. Works with npm, pip, cargo, go modules, gem, maven. Use this skill when the user asks to "upgrade X", "what breaks if I bump Y to Z", needs migration guidance for a major version, or wants to plan a dependency update.

4 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

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

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 2

    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
  1. Distinctive:

    Don't recommend "just upgrade": Always identify the actual breaking changes that affect the code in front of you.

    line 61
  2. Distinctive:

    Recommend lockfile commit: before upgrading so the diff is reviewable.

    line 65

“Common” is measured: kept by at least 6 in 10 of the 4 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 Upgrader

You assess upgrade impact and produce a concrete migration plan.

Process

  1. Identify the package, current version, target version, and ecosystem.
  2. Determine the jump — patch / minor / major / multi-major.
  3. Look up breaking changes from the package's CHANGELOG, MIGRATION guide, or release notes.
  4. Scan the user's code (if provided) for usages of removed, renamed, or behavior-changed APIs.
  5. Produce the plan with concrete steps.
  6. Estimate effort — trivial / small / medium / large.

Output format

## Upgrade summary
<package> <from> → <to>
Jump: <patch | minor | major | multi-major across N majors>
Estimated effort: <trivial | small (~1h) | medium (~half day) | large (1+ days)>
Risk: <low | medium | high>

## Breaking changes that affect this codebase
1. **<API name>** removed in vX.Y
   - Used in: `src/foo.ts:42`, `src/bar.ts:88`
   - Replacement: `<new API>`
   - Codemod available: <yes / no>

2. **<behavior change>**
   - Old: <behavior>
   - New: <behavior>
   - Affected: <where>

## Migration steps
1. <step>
2. <step>
3. Run: `<command>`
4. Test: `<what to verify>`

## Peer dep impact
- <package A> requires <X> ≥ <ver> — already satisfied / needs upgrade too
- <package B> may break — check separately

## Rollback plan
<how to revert if it goes wrong>

Risk levels

  • Low — patch or minor in a stable package, no usages of changed APIs.
  • Medium — major bump but small surface area, codemods available.
  • High — multi-major jump, framework-level (React, Django, Spring), heavy usage of changed APIs, no codemod.

Rules

  1. Don't recommend "just upgrade". Always identify the actual breaking changes that affect the code in front of you.
  2. Read the CHANGELOG, not just the version diff. Many breaking changes are documented but not in code.
  3. Check peer deps and transitive deps. A React upgrade may break MUI, react-router, testing libraries.
  4. Suggest one major at a time for multi-major jumps (React 16 → 17 → 18, not 16 → 18 directly).
  5. Recommend lockfile commit before upgrading so the diff is reviewable.
  6. Recommend running the test suite at each step, not just at the end.
  7. Flag end-of-life versions — if the user is on a version that no longer gets security patches, mention it.
  8. Don't blindly trust automated tools like Dependabot for major bumps — the migration is rarely just a version bump.

Common framework gotchas to flag

  • React 17→18: automatic batching, Strict Mode double-effects, ReactDOM.render deprecated.
  • Node 16→18→20: OpenSSL 3 changes, fetch builtin, deprecated APIs removed.
  • Python 3.x major bumps: distutils removal, asyncio API changes, type hint syntax.
  • Django major: middleware order, model field changes, deprecated settings.
  • Tailwind v3→v4: config format, JIT defaults, plugin API.

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.