Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add tmj-90/gaffer --skill changelog-generatorgit clone --depth 1 https://github.com/tmj-90/gafferWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/tmj-90/gaffer/changelog-generator)<a href="https://agentmods.dev/skills/tmj-90/gaffer/changelog-generator"><img src="https://agentmods.dev/badge/skills/tmj-90/gaffer/changelog-generator/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tmj-90/gaffer/changelog-generator"><img src="https://agentmods.dev/badge/skills/tmj-90/gaffer/changelog-generator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00068 | $0.01280 |
| Opus 5.5 | $0.00027 | $0.00512 |
| Sonnet 5.5 | $0.00014 | $0.00256 |
| Haiku 4.5 | $0.00007 | $0.00128 |
Grade A, and why
changelog-generator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 27d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generate consistent, auditable release notes
A changelog is for humans deciding whether and how to upgrade — "don't let your friends dump git logs into changelogs" (Keep a Changelog 1.1.0). Commits are the input; the output is a curated list of user-visible changes plus a version number that follows Semantic Versioning 2.0.0.
Rules of the formats
Conventional Commits 1.0.0 → SemVer bump (take the highest in the range):
| Signal | Bump |
|---|---|
! before the colon (feat(api)!:) or a BREAKING CHANGE: / BREAKING-CHANGE: footer (uppercase) |
MAJOR |
feat: |
MINOR |
fix: |
PATCH |
any other type (docs, refactor, perf, test, build, ci, chore, …) |
none by the spec; many tools treat perf as PATCH — follow the repo's config |
Types are case-insensitive except BREAKING CHANGE. While the version is 0.y.z,
SemVer allows anything to change; most repos bump MINOR for breaking changes in 0.x —
follow the repo's history or release config.
Keep a Changelog 1.1.0: newest version first; an ## [Unreleased] section at the
top; each release ## [1.4.0] - 2026-06-28 (ISO 8601 date); only the sections that have
entries, from Added, Changed, Deprecated, Removed, Fixed, Security; comparison
links at the bottom ([1.4.0]: https://.../compare/v1.3.0...v1.4.0); a pulled release is
marked ## [1.3.1] - 2026-05-02 [YANKED].
Procedure
- Find the convention. Read the existing
CHANGELOG.md, release config (.releaserc,release-please-config.json,cliff.toml,.changeset/) anddocs/RELEASING.mdor equivalent;search_lorefor release conventions. Match the existing heading, date and link style exactly. If a release tool owns the file, change its input (config, changeset) rather than hand-editing its output. - Get the range. Last release tag to
HEAD:git describe --tags --abbrev=0thengit log --no-merges --format='%H%x1f%s%x1f%b%x1e' <tag>..HEAD. Include bodies — breaking-change footers live there. - Classify each commit. Parse type, scope,
!and footers. If commits are not Conventional (Gaffer's own aredeliver #N: …), classify from the diff and the linked ticket instead of rejecting them — never rewrite history to fix messages. Drop a commit and itsrevert:when both are in the range. Drop pure internal changes (tests, CI, refactors with no behaviour change) from the notes. - Write entries for users. One line per user-visible change, in the imperative or past tense consistently, naming the feature/endpoint/flag, with the PR or issue reference the repo uses. Map: new capability → Added; changed behaviour → Changed; scheduled removal → Deprecated; removed → Removed; bug fix → Fixed; vulnerability fix → Security. Prefix breaking entries with BREAKING: and include the migration step.
- Place the entry. Unreleased work goes under
## [Unreleased]. Only when the ticket asks to cut a release: move those entries under the new version heading with today's date, add the compare link, and state the computed version. You never create or push tags — the release pipeline does. - Verify (below) and evidence the acceptance criteria with the
record-evidenceskill (diff_summaryfor the changelog diff, plus the classification table as proof of the bump), then stop.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 27d ago Changed · -14 lines 17f857f47daf
- 1mo ago First seen · 103 lines · 68 tokens per session scan A e8fb1dd812d8
changelog-generator is a skill published in the GitHub repository tmj-90/gaffer (2 stars, last pushed 28d ago), licensed Apache-2.0. It adds 68 tokens to every session and 1,280 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
besimple-tiny-broccoli
Small-change wrapper: implement → run repo checks → atomic commit → run dedup (BASESHA..HEAD).
code-review-loop
Iterative review+fix loop for BASESHA..HEAD: generate findings, apply accepted fixes, run checks, commit, and re-review up to 3 iterations or until clean.
dedup
Dedupe-only pass for BASESHA..HEAD: remove duplicate code introduced by the diff or reuse existing shared utils; applies changes + commits.
plan-critique-loop
Critique and revise an existing plan doc up to 3 iterations, using accept/reject triage and stopping early when no important feedback remains. Use when refining a plan/.md before implementation.
claude-simplify-wrapper
Run Claude's built-in /simplify skill on BASESHA..HEAD, validate checks, and commit.
dev-ship
Fechamento de feature — verificação goal-backward antes de declarar pronto. Roda Must-Haves do PLAN.md, executa o demo script, revisa o diff completo procurando bugs e restos (debug logs, TODOs, código morto), passa lente de segurança nos arquivos tocados, escreve SUMMARY.md e arquiva o plano. Use quando o usuário…