Release change analyzer

Skill ArabelaTso/Skills-4-SE/skills/release-change-analyzer

Compares HEAD with the latest published version to analyze real changes, group by type, and recommend version bumps. Use before publishing a release to understand what actually changed.From its SKILL.md

Install

npx -y skills add ArabelaTso/Skills-4-SE --skill release-change-analyzer

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

2.7 KB, 613 tokens by cl100k_base, as published. Nobody here has run it

Release Change Analyzer

Analyzes unpublished changes by comparing HEAD with the latest published version. Reads actual diffs instead of just copying commit messages.

When to Use This Skill

Before publishing a new release
Deciding between major/minor/patch version bump
Writing accurate release notes based on real code changes
Reviewing what changed since last deployment

What This Skill Does

Step 1: Identify Published Version

# npm
npm view <package-name> version 2>/dev/null || echo "not published"

# PyPI
pip index versions <package-name> 2>/dev/null | head -1

# Or use git tags
git tag --sort=-v:refname | head -1

Step 2: Analyze Real Changes

# Get actual diff (NOT just commit messages)
git diff v{published-version}..HEAD --stat
git diff v{published-version}..HEAD

For each commit, MUST:

Read the actual diff to understand WHAT CHANGED
Describe the REAL change in plain language
Explain WHY it matters (if not obvious)

Step 3: Group and Report

## Unpublished Changes (v{published} -> HEAD)

### Features
| Scope | What Changed |
|-------|--------------|
| auth | Added OAuth2 support with Google and GitHub providers |

### Fixes
| Scope | What Changed |
|-------|--------------|
| api | Fixed race condition in concurrent request handling |

### Refactoring
| Scope | What Changed |
|-------|--------------|
| core | Extracted retry logic from 3 services into shared module |

### Breaking Changes
- Removed deprecated `v1/login` endpoint

### Files Changed
42 files changed, 1205 insertions(+), 387 deletions(-)

### Suggested Version Bump
- **Recommendation**: minor
- **Reason**: New features added, no breaking changes in public API

Version Bump Decision

Condition	Bump
Breaking changes in public API	major
New features, no breaking changes	minor
Bug fixes, refactoring, docs only	patch

Anti-Patterns

Copying commit messages verbatim — read the actual diff
Describing "what" without "why" — explain impact
Missing breaking changes — always check public API surface
Guessing at scope — verify from the diff

Example

User: "What changed since last release?"

Output: Reads diff between v1.2.3 and HEAD, identifies 3 features, 2 fixes, 1 refactor, no breaking changes. Recommends minor bump to v1.3.0.

Inspired by: oh-my-opencode get-unpublished-changes command

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most ship operate skills give in 613 tokens

Counted across 779 of the 1,178 authors here whose files we hold, read 2026-08-07

Document a rollback plan before deploymentin 41 of 779, across 22 files
Update the changelogin 21 of 779, across 19 files
Run the test suitein 20 of 779
Create an annotated git tagin 20 of 779
Clean up feature flags after full rolloutin 18 of 779, across 10 files
Verify deployment health after launchin 18 of 779, across 10 files
Test both feature flag statesin 17 of 779, across 9 files
Verify the working tree is cleanin 17 of 779
Make database migrations backward-compatiblein 16 of 779, across 8 files
Set up error monitoring before launchin 15 of 779, across 7 files
Monitor metrics at each rollout stagein 14 of 779, across 5 files
Create a GitHub releasein 14 of 779

Said here and by no other author read

identify the latest published version
compare head with the latest published version
describe real changes in plain language
explain why changes matter
check the public api for breaking changes

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.