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 atman-33/workhub --skill release-appgit clone --depth 1 https://github.com/atman-33/workhubWrote 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/atman-33/workhub/release-app)<a href="https://agentmods.dev/skills/atman-33/workhub/release-app"><img src="https://agentmods.dev/badge/skills/atman-33/workhub/release-app/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/atman-33/workhub/release-app"><img src="https://agentmods.dev/badge/skills/atman-33/workhub/release-app.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 | $0.01172 |
| Opus 5 | $0.00034 | $0.00586 |
| Sonnet 5 | $0.00014 | $0.00234 |
| Haiku 4.5 | $0.00007 | $0.00117 |
Grade A, and why
release-app 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 5d 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 — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
release-app — Publish a workhub release
A workhub release is a tag push, not a PR: pushing vX.Y.Z to main is
what builds and publishes the GitHub Release. That makes the push the point of
no return — a published version is never re-tagged — so this skill front-loads
every check and stops for approval before it.
The release contract (tag format, required assets, hard invariants) is not restated here. It lives in the repository, and step 1 reads it.
1. Read the contract
Read .claude/rules/release-process.md in the workhub repository. It is the
authority for the tag format, the asset names the release must carry, and the
invariants that must never break. Everything below defers to it.
Completion criterion: you can name the required tag format, every required release asset, and what breaks if one is renamed.
2. Get onto a clean, current main
git status --short
git checkout main
git pull --ff-only origin main
An unclean tree is a stop, not a thing to tidy: show the user what is there and let them decide. Never stash, reset, or discard on their behalf.
Completion criterion: git status --short is empty and local main equals
origin/main.
3. Check that main is green
gh run list --branch main --limit 1
Anything other than a completed success is a stop — report which run and why.
A red main that gets tagged burns a version number.
4. Decide what is being released
git describe --tags --abbrev=0 # last released version
git log <last-tag>..HEAD --oneline --no-merges
Read the current version from src-tauri/Cargo.toml and compare it with the
last tag. Two states are normal:
- Version already bumped, tag missing. A previous run prepared the version and stopped. Reuse it — unless commits landed after the changelog entry was written, in which case those commits still need representing.
- Version equals the last tag. A fresh bump is needed. Propose it from what the commits actually are: breaking → major, feature → minor, fix → patch.
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.
- 5d ago First seen · 137 lines · 68 tokens per session scan A 4bc365e0bba1
release-app is a skill published in the GitHub repository atman-33/workhub (2 stars, last pushed 3d ago), licensed MIT. It adds 68 tokens to every session and 1,172 once invoked, about $0.0003 per session on Opus 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 2026-09-03.
Other skills, from other repositories
ship
Ship a build — pre-flight checks, deploy, document, verify, monitor. Use when deploying code, publishing packages, or releasing.
release
Prepare a version release — bump version files, commit, and tag. Just run /release with no arguments.
release-plugin
Cut a cairn plugin release for a given MAJOR.MINOR.PATCH version, running every pre-flight and version gate before anything is changed or published.
release-init
Detect project type and generate a tailored project-level /release skill. Run once per project to set up releasing.
release-please-monorepo
Set up automated semantic versioning, changelogs, and per-package publishing in a pnpm/Turbo monorepo using googleapis/release-please-action v4. Use when the user asks to "add release please", "automate releases/versioning", "publish packages on push to main", "set up release-please in a monorepo", or wants…
id-rollback
The infra-delivery INCIDENT safe-revert PROCEDURE skill (the recurring/incident rollback step) — the PROCEDURE the generic agent runs when a delivered change did NOT land (a NOT-LANDED realization, an escaped defect, bad config) to restore the last-known-good IaC and prove reality matches it again. The shape is git…