dependency-upgrade
Skill1 starsChanged 4 months agoMITView on GitHub
Use this skill when asked to upgrade an existing package, bump dependencies, update a lockfile, or migrate across a major version. Triggers include: "upgrade <package>", "bump dependencies", "update the lockfile", "migrate to v<N>", "update packages to latest". Do NOT use this skill for installing a brand-new dependency that isn't already in the project — that's a plain install, not an upgrade, and carries none of the breaking-change risk this skill exists to gate.
23 other projects use a version of this.
Adopted by
Most recently taken up by
- paureis/pau-skillstheir own version, first seen Nov 8, 2026
- HyperVon/agent-guidance-kittheir own version, first seen Nov 1, 2026
- navikt/copilottheir own version, first seen Nov 1, 2026
- agentkitco/agentkit-dev-skillstheir own version, first seen Nov 1, 2026
- Eliahhango/nullbyte-securitytheir own version, first seen Nov 1, 2026
- itsofficialadi/boundifftheir own version, first seen Oct 25, 2026
Used with
Projects with it also use incident-response, code-review and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
AnthropicSkillincident-response
Run an incident response workflow — triage, communicate, and write postmortem. Trigger with "we have an incident", "production is down", an alert that needs severity assessment, a status update mid-incident, or when writing a blameless postmortem after resolution.
Also in 3 projects that use this
GoogleSkillcode-review
Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.
Also in 6 projects that use this
GoogleSkilltdd
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
Also in 4 projects that use this
MicrosoftSkillhandoff
Compact the current conversation into a handoff document for another agent to pick up.
Also in 3 projects that use this
- +1 more from google
Other versions23 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 23
- secondsky/sap-skills★ 462same name
- navikt/copilot★ 61same name
- xfstudio/skills★ 9same name
- KhaledSaeed18/dotclaude★ 5same name
- Eliahhango/nullbyte-security★ 4same name
- juanitourquiza/shipframe★ 4same name
- WST2002/claude-code-skills★ 4same name
- 0disoft/mcp-security-proxy★ 2same name
- ai4brands-design/claude-skills★ 2same name
- CodeWithBehnam/cc-docs★ 2same name
13 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- flaky-test-fixerSkill
- frugal-contextSkill
- incident-triageSkill
- root-cause-debuggerSkill
- run-receiptSkill
- session-handoffSkill
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 22 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
- Distinctive: line 48
Never bump more than one major version in a single step (e.g.
- Distinctive: line 51
Never skip the test gate, even for a patch/minor bump that "should" be safe — that assumption is exactly what this skill exists to not make.
- Distinctive: line 53
Always leave a rollback path stated in exact, runnable commands, not "you could probably just revert this."
“Common” is measured: kept by at least 6 in 10 of the 22 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.
Files
Dependency Upgrade
When this fires
Upgrades are usually done ad hoc: bump the version, see what breaks, patch around it. That works until a major-version jump silently changes behavior that isn't caught by whatever happened to be run afterward, or until a multi-package bump makes it impossible to tell which bump broke what. This skill locks the upgrade into a bounded, gated, reversible sequence.
Protocol
- Read the changelog / migration notes for the exact version delta before touching anything — not a summary from memory, the actual changelog or migration guide for the specific version jump being made (current version → target version).
- Bump one package at a time. Even if multiple packages need upgrading, each gets its own bump-gate-commit cycle. Never bundle unrelated package bumps into a single step — it destroys the ability to attribute a failure to a specific change.
- Run the test gate after each bump, before making any other change. The gate is whatever the project already uses to verify correctness (test suite, build, typecheck) — run it, don't assume it would pass.
- On gate failure, apply only the migrations documented in step 1's changelog/migration guide, then re-run the gate. Do not improvise fixes for breaking changes that aren't in the documented migration path — if the failure isn't explained by the documented changes, stop and investigate before proceeding (it may not be a migration issue at all).
- Produce a rollback plan — the exact commands to revert this specific
bump (e.g.
git revert <sha>, or the exact prior version pin) — before considering the upgrade finished, whether or not anything went wrong.
Hard rules
- Never bump more than one major version in a single step (e.g. v2 → v4 must go through v3 first, or be explicitly called out as a deliberate skip with its own migration research, never assumed safe).
- Never skip the test gate, even for a patch/minor bump that "should" be safe — that assumption is exactly what this skill exists to not make.
- Always leave a rollback path stated in exact, runnable commands, not "you could probably just revert this."
- Migrations applied must trace back to the documented changelog/migration guide read in step 1 — don't guess at what a breaking change wants.
Output
Per package bumped:
Package: <name> <old version> -> <new version>
Changelog/migration notes reviewed: <link or source>
Gate before bump: <pass/fail — baseline>
Gate after bump: <pass/fail>
Migrations applied: <none, or the specific documented changes made>
Gate after migration: <pass/fail>
Rollback plan: <exact command(s) to revert this bump>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
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.

