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 petekp/claude-code-setup --skill release-notesgit clone --depth 1 https://github.com/petekp/claude-code-setupWrote 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/petekp/claude-code-setup/release-notes)<a href="https://agentmods.dev/skills/petekp/claude-code-setup/release-notes"><img src="https://agentmods.dev/badge/skills/petekp/claude-code-setup/release-notes/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/petekp/claude-code-setup/release-notes"><img src="https://agentmods.dev/badge/skills/petekp/claude-code-setup/release-notes.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 | $0.00046 | $0.01117 |
| Opus 5.5 | $0.00018 | $0.00447 |
| Sonnet 5.5 | $0.00009 | $0.00223 |
| Haiku 4.5 | $0.00005 | $0.00112 |
Grade A, and why
release-notes 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 yesterday.
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.
Release notes
Write for someone who already uses the product. They want to know what will feel different after they update, and whether they have to do anything. Most read the first line and stop. Some read the next section. Few read further.
Gather
- List the changes since the last release, such as with
git log --oneline <last tag>..HEAD. Read the PRs and issues they reference. - Read a change's diff when its commit subject doesn't make the user's experience clear. Describe only effects you can trace to a change.
- Find the project's update steps and the apps or platforms it ships to. Look in its release doc, its README, and the last release's notes.
Shape
In this order:
- One line on the effect users will notice most. Name the effect, not the area: "Other mods' notes come back above your prompt", not "Fixes for the band".
- A picture of the headline change when it is visible. Reuse the before-and-after image from the PR that made the change when there is one.
- What you'll notice. One item per change, as described below.
- Update. The project's update steps, as the last release gave them, changed only where this release changes them.
- Under the hood. One line per change that a technical user or contributor would want to know: the mechanism behind a change above, a fix with no visible effect, a changed file format or API. Link the issue or PR when there is one, and thank the person who reported it. Work that does not ship in the product, such as a video, a plan, or a refactor with no effect, stays out.
When nothing in a release is visible, say so in the first line, and keep only Update and Under the hood.
What you'll notice
- Start each item with a bold sentence that says what the user sees or does now. Follow it with one sentence on how it was before, when that helps them recognize the change.
- Name the app or platform when a change reaches only some of them, such as "Codex:". When an app has no changes, say "Nothing changed in Claude Code."
- Put anything the user must do inside the item it belongs to, such as "quit and reopen the app once". A step that applies to the whole release goes in Update.
- Use the names users see on screen: button labels, tab names, and messages, quoted exactly. Name the place in plain words, such as "the terminal" or "the Codex tab". Code names and the project's own terms for its parts belong under the hood.
- Order items by how often users run into them. A change everyone sees comes before an edge case. A removed feature or a change that breaks a workflow comes first, with what to do instead.
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.
- yesterday First seen · 70 lines · 46 tokens per session scan A 7898ad44056f
release-notes is a skill published in the GitHub repository petekp/claude-code-setup (48 stars, last pushed yesterday), licensed MIT. It adds 46 tokens to every session and 1,117 once invoked, about $0.0002 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 2026-10-22.
Other skills, from other repositories
publish-site
Versioned site deploys to GitHub/Cloudflare/Netlify Pages.
github
GitHub via gh CLI: PRs, issues, reviews, repos, auth.
baby-sit
Monitor a GitHub pull request until CI is green, diagnose failures, and rerun only evidence-backed flaky GitHub Actions jobs.
changelog
Turn a range of commits or merged PRs into a changelog entry grouped by change type. Use when the user asks for release notes, a changelog, or "what changed" between two points.
otto-drive-pr
PR mergeability, CI, and live preview.
shiplog
Recap of everything shipped since the last run - cross-repo PRs, security fixes, star deltas, and X traction, synthesized into a digest article and a ready-to-post shiplog in your voice.