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 agentmods add skills/genkovich/sdd/shipnpx skills add genkovich/sdd --skill shipgit clone --depth 1 https://github.com/genkovich/sddWrote 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/genkovich/sdd/ship)<a href="https://agentmods.dev/skills/genkovich/sdd/ship"><img src="https://agentmods.dev/badge/skills/genkovich/sdd/ship.svg" alt="Measured on agentmods" 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 | $0.00154 | $0.01422 |
| Opus 5 | $0.00077 | $0.00711 |
| Sonnet 5 | $0.00031 | $0.00284 |
| Haiku 4.5 | $0.00015 | $0.00142 |
Grade A, and why
ship 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: ship
The closing step. review confirmed the change is correct on paper; ship confirms it works in reality and packages it for merge. The loop ends here: a reviewed, verified change with a changelog and an open PR — not a merge to main (that stays a human decision).
Forge-agnostic and stack-agnostic: the verification commands are detected the way implement detects them; the PR step targets whatever forge the remote points at (GitHub via gh, GitLab via glab, or copy-paste).
Changelog + PR-body prose follow artifact_language — commit messages, branch names and the SDD-Task/SDD-AC trailers stay English → ../_shared/artifact-language.md.
Owner
The implementer (drives) + the reviewer who signed off in review.
Inputs
<slug>— feature slug.- Gate (hard refuse): a
PASSreview record (docs/features/<slug>/_review/) or, at minimum, an implemented + gate-green change. No review yet → «runreview <slug>first». - Read:
spec.md(what to claim in the changelog), Acceptedadr/(decisions worth recording), the feature's commits (theSDD-Taskhistory).
Protocol
- Final verification — does it actually work. Re-run the detected gate (unit + integration where available + lint + vet). Then run the feature for real against its acceptance criteria — not just "tests pass": start the app / hit the endpoint / exercise the flow and observe the spec's outcomes (e.g. the default-on read returns defaults; an invalid value is rejected). Concretely: spot-check at least 3 of the most critical §5 AC outcomes (fewer only if the spec has fewer; scale the count with the feature's breadth), and for each name the AC id + the behaviour actually observed — «AC-03: posted the same apply twice → one discount row» — so the verification is checkable, not a vibe. If the app can't be run here (no runtime, no Docker), say so explicitly and record what was verified vs deferred — never claim verified-working when only tests compiled.
- Write the changelog / KB note. From
./templates/changelog.md: what changed, why (link spec + the key ADRs), any migration/operational note (e.g. "adds migration 000023 — run it on deploy"), and how to use it. Partner-facing if the change is partner-facing. - Prepare the PR. Ensure the work is on a feature branch (not the default branch). Draft the PR body from
./templates/pr-body.md: summary, the AC it satisfies, links to spec/sad/ADRs, theSDD-Taskcommit list, the test + verification evidence, and any migration/rollback note. - Detect the forge + propose the PR command. Inspect the remote:
github.com→gh pr create;gitlab.com/self-hosted GitLab →glab mr create; otherwise print the branch + body for manual creation. Propose the command — do not run a push/PR to a shared remote without the user's go-ahead, and never merge to main. - Update the roadmap. In
docs/roadmap.md(viaroadmap) set the step'sStatus: shippedand add the date + PR/changelog link to Shipped. This is the anti-drift hook: delivery itself keeps the roadmap current. (No roadmap yet → skip; it's optional.) - Summary (terminal handoff). Emit the stage-handoff block per
../_shared/handoff.md(terminal variant) — What I did (verification result: verified-working / what was deferred and why; the roadmap update) + Review (the changelog path + the PR) + Run next = Done: the PR command (or URL if the user ran it) — merging to main is your call; there is no/sddsuccessor.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 First seen · 61 lines · 154 tokens per session scan A 43b9e574f6f6
ship is a skill published in the GitHub repository genkovich/sdd (118 stars, last pushed 15d ago), licensed MIT. It adds 154 tokens to every session and 1,422 once invoked, about $0.0008 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 2026-08-30.
Other skills, from other repositories
release
Cut a Symphony release by bumping the committed version, landing it, tagging the merged commit, and verifying the Burrito release workflow. Use when asked to release, tag, or retag Symphony.
changelog-entry
Add a new entry to CHANGELOG.yml under the current unreleased version (or create the version block if needed), then regenerate documentation. Use when the user says things like "add a changelog entry", "log this fix in the changelog", or "/changelog-entry".
release-notes
Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.
mate-oss-gate
在准备把 MateCloud（或其子集）开源 / 发布到公开仓前使用。按 open-core 边界把关：剥离企业代码、清竞品名与内部路径、查密钥、确认 LICENSE。当用户说"要开源了""发布公开版""开源前检查""oss release"时触发。.
devops/changelog-generation
自动生成 CHANGELOG，基于 git 提交历史和 pipeline 产物信息，遵循 Conventional Commits 和 Keep a Changelog 规范.
changelog-composer
Generates structured changelogs and release notes from git history and PRs, classifying breaking changes, features, fixes, performance, docs. Triggers on: "generate changelog", "write release notes", "what changed since", "prepare release", "release notes for", "diff since tag".