Skip to content

JPeetz/agent-skills / skills/git-release/changelog-generator

changelog-generator

Verified

Use this skill when cutting a release, generating CHANGELOG.md from commits, computing next semver version, or linting commit messages. Parses Conventional Commits to detect breaking changes, render Keep a Changelog sections, and enforce commit format in CI.

FileKindSize
CHANGELOG.mdmarkdown673 B
LICENSEother1.1 KB
SKILL.mdskill md7.1 KB
evals/evals.jsonconfig3.9 KB
references/ci-integration.mdmarkdown6.6 KB
scripts/commit_linter.pyscript10.6 KB
scripts/generate_changelog.pyscript15.1 KB

evals/evals.json

View on GitHub
{
  "skill_name": "changelog-generator",
  "evals": [
    {
      "id": 1,
      "prompt": "We're on v1.4.0. Since that tag we have: feat(api): add pagination, fix: timezone drift, docs: pagination params, and a commit 'feat(api)!: drop the v1 tasks endpoint' whose body has 'BREAKING CHANGE: clients must migrate to GET /v2/tasks.' What version is next and what does the changelog entry look like?",
      "expected_output": "Version 2.0.0 (a breaking change forces a major bump, not 1.5.0). A Keep a Changelog entry dated for the release with Added (pagination), Fixed (timezone drift), Documentation (params), Changed containing the dropped endpoint, and a migration note quoting the BREAKING CHANGE footer. The docs commit may be included; there are no chore commits to suppress in this input.",
      "expectations": [
        "Bumps to 2.0.0, not 1.5.0 or 1.4.1",
        "Breaking change appears under Changed, not Added",
        "Migration note from the BREAKING CHANGE footer is preserved verbatim",
        "Groups commits into Added / Changed / Fixed sections rather than listing raw commit lines"
      ]
    },
    {
      "id": 2,
      "prompt": "Generate the changelog entry for the range since v1.4.0 and save it into our existing CHANGELOG.md, which already has a 1.4.0 section.",
      "expected_output": "Run scripts/generate_changelog.py --from v1.4.0 --version <next> --update CHANGELOG.md. The new section is prepended above the existing 1.4.0 section, the existing history is untouched, and the script refuses (non-zero exit) if that version is already present.",
      "expectations": [
        "Uses the --update flag with --from and --version",
        "New entry goes above existing sections, not replacing them",
        "Mentions the duplicate-version guard as the failure mode"
      ]
    },
    {
      "id": 3,
      "prompt": "Our git history is full of messages like 'fixed stuff', 'WIP', 'update code', and 'Merge branch main into feature'. Can I generate a changelog from this?",
      "expected_output": "Not usefully. The generator skips merge commits and non-Conventional messages, so those commits produce no entries. Run scripts/commit_linter.py over the history first to see which commits are unparseable, adopt the Conventional Commit format going forward, and only then rely on generation. Do not hand-write entries that pretend the commits were structured.",
      "expectations": [
        "States that non-conventional and merge commits are skipped, not coerced",
        "Recommends the linter as the enforcement step before generation can be trusted",
        "Does not invent changelog lines for unparseable commits"
      ]
    },
    {
      "id": 4,
      "prompt": "I want to block badly formatted commit messages in CI before they merge. How do I wire up the linter?",
      "expected_output": "Run scripts/commit_linter.py --range origin/main..HEAD (or the PR base..head range) in a CI step; it exits 1 on violations and 0 when clean, and --format json gives a machine-readable report. Use --strict to also fail on style warnings such as trailing periods or uppercase descriptions.",
      "expectations": [
        "Uses --range with a PR-scoped range rather than the whole history",
        "Notes the exit-code contract (1 = violations, 0 = clean)",
        "Mentions --strict and/or --format json for CI reporting"
      ]
    },
    {
      "id": 5,
      "prompt": "Since the last tag we only have 'chore: bump dev deps' and 'test: add coverage for parser'. Should I cut a patch release?",
      "expected_output": "No. Both types are omitted from the changelog, so the commit stream implies no bump. Skip the release; cutting a patch version for internal-only changes teaches consumers that version numbers are noise.",
      "expectations": [
        "Declines the release rather than suggesting a patch bump",
        "Explains that chore/test commits are non-releasable by type",
        "Notes the cost of releasing with no consumer-visible change"
      ]
    }
  ]
}
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall