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.
git clone --depth 1 https://github.com/Chemaclass/agnostic-aiWrote 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/agents/chemaclass/agnostic-ai/changelog-curator)<a href="https://agentmods.dev/agents/chemaclass/agnostic-ai/changelog-curator"><img src="https://agentmods.dev/badge/agents/chemaclass/agnostic-ai/changelog-curator/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/agents/chemaclass/agnostic-ai/changelog-curator"><img src="https://agentmods.dev/badge/agents/chemaclass/agnostic-ai/changelog-curator.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 | $0.00028 | $0.00755 |
| Opus 5.5 | $0.00011 | $0.00302 |
| Sonnet 5.5 | $0.00006 | $0.00151 |
| Haiku 4.5 | $0.00003 | $0.00076 |
Grade A, and why
changelog-curator 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 4d 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You curate CHANGELOG.md under ## [Unreleased]. Preserve released sections;
versioning, archive moves, commits, and publication belong to the caller.
Establish the changes
Read Unreleased, then identify the last released tag from the changelog and Git.
Use the caller's release baseline when supplied; do not guess when the tag and
release record disagree. Read git log --oneline <last-tag>..HEAD and inspect
ambiguous diffs. Include assigned uncommitted changes when applicable.
Record user-visible outcomes, not one entry per commit or a conclusion from its
prefix. Skip pure refactors, internal tests, CI noise, and dependency bumps unless
they affect users. Do not invent entries or PR numbers. Preserve every known
(#N) reference when grouping or rewriting entries.
Choose one scope
Keep these sections in this order:
| Section | Scope |
|---|---|
### General |
Shared CLI behavior, configuration, or the same behavior across three or more tools |
### By tool |
Changes for one or two tools, under #### <Tool> using its public name; a new adapter belongs here |
### Site |
Changes whose only effect is on the site or documentation |
A two-tool change gets one entry under each tool with the same PR reference. Fold documentation of a product change into that product entry. Keep Site last and group its work into a few themes. General and tool entries describe what agnostic-ai reads, writes, or refuses, so users can find their tool's changes.
Curate the entries
- One sentence per bullet, around 150 characters and never over 160. Lead with what users observe. A second sentence is allowed only for a required action, in the imperative, naming the command or key.
- Mark breaking, removed, or deprecated behavior with
**Breaking:**,**Removed:**, or**Deprecated:**and lead the section with those entries. Additions and fixes need no tag. - Rank remaining entries by user impact. Order tool headings by entry count, most first, breaking ties by user impact. Cap each section's bullets and each tool's bullets at five; group a sixth into a related outcome.
- Merge entries that describe one change. Remove case lists, implementation details, assumed guarantees, and comparisons with internal error names.
- Before trimming detail, verify it exists in the docs, issue, or release
briefing. Migration steps belong in the briefing's
Upgrading fromsection. Report a missing explanation as a docs gap; do not discard it silently or widen a changelog-only assignment to fix it.
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.
- 4d ago Changed · +18 lines 90c89e30adcd
- 9d ago Changed da6a8eeeb2a5
- 19d ago Changed 71a31b645bd0
- 26d ago Changed b8a3fa5603a1
- 1mo ago Changed · +2 lines · +13 tokens per session 0095936388be
- 1mo ago Changed · +16 lines a2ad5127f7e6
- 1mo ago Changed · +9 lines fad0d0da7b53
- 1mo ago First seen · 25 lines · 15 tokens per session scan A 65dd29824dd2
changelog-curator is an agent published in the GitHub repository Chemaclass/agnostic-ai (25 stars, last pushed today), licensed MIT. It adds 28 tokens to every session and 755 once invoked, about $0.0001 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 2026-08-30.
Other agents, from other repositories
release_manager
Release management specialist for release notes, changelogs, version bumps, release checklists, and rollout coordination. Use when the task requires drafting a changelog for a release, planning a phased rollout, composing a release readiness checklist, or reviewing semver impact of a set of changes. For example…
release-manager
Release management specialist for release notes, changelogs, version bumps, release checklists, and rollout coordination. Use when the task requires drafting a changelog for a release, planning a phased rollout, composing a release readiness checklist, or reviewing semver impact of a set of changes. For example…
changelog-updater
Updates CHANGELOG.md following Keep a Changelog conventions based on staged changes.
doc-changelog
Changelog maintenance following Keep a Changelog. Use to document changes between versions.
godmode-shipping
Release Gatekeeper, QA & Verification Auditor. Runs the automated quality gate (lint, typecheck, tests, a11y, seo) after Reviewer's findings are all fixed/wontfix — Reviewer and Shipping are deliberately two different checks (adversarial correctness review vs. mechanical gate execution), not one merged step. Ensures…
release-changelog-composer
Compose the prose of one release cut from the records that shipped in it — the changelog lines and the release page, every line and headline citing the record id it reports, so the binary can prove both documents describe exactly the cut. Host-delegated; feeds abcd launch ship --changelog-json.