Borrowing it
Nothing to install: this file belongs to Cratis/VerticalSlices. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Cratis/VerticalSlices/main/.ai/skills/ship-changes/SKILL.mdgit clone --depth 1 https://github.com/Cratis/VerticalSlicesWrote 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/cratis/verticalslices/ship-changes)<a href="https://agentmods.dev/skills/cratis/verticalslices/ship-changes"><img src="https://agentmods.dev/badge/skills/cratis/verticalslices/ship-changes/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/cratis/verticalslices/ship-changes"><img src="https://agentmods.dev/badge/skills/cratis/verticalslices/ship-changes.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 | $0.00072 | $0.03770 |
| Opus 5.5 | $0.00029 | $0.01508 |
| Sonnet 5.5 | $0.00014 | $0.00754 |
| Haiku 4.5 | $0.00007 | $0.00377 |
Grade A, and why
ship-changes 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 1mo 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 — 352 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ship Changes
This skill takes local modifications from the working tree and lands them on
main via a properly structured branch, commits, and PR — following all
project conventions for commits, PR descriptions, and labels — and leaves no
resolved issue open behind it.
Never rewrite history while shipping. No
rebase, nocommit --amend, noreset --hard, nopush --force(or--force-with-lease), no deleting a branch that still holds unmerged commits — on any branch, including your own, unless the human asks for that exact command in that exact message. Correct a mistake with a new commit; undo withgit revert; move work withgit cherry-pick; integrate withgit merge. Do not tidy a messy series before opening the PR — the series is the record. See git-commits.md for the full prohibition and the recovery procedure.
Inputs
Collect the following before starting:
- Semantic label —
patch,minor,major, or none. Skip labeling entirely if the user says no label or omits the label. Do not ask — infer from the nature of the change or follow the user's explicit instruction. - Branch name suffix — short kebab-case description of the work, e.g.
fix/testing-orleans-runtime-assemblies. Determine from the nature of the changes if not provided. - Related GitHub issue — search GitHub issues if the change likely relates to one; use the real number or omit the reference when none exists. Never invent or reuse example numbers. Keep the numbers you find — they are needed twice: as the
(#N)reference in the PR bullets, and to close the issues after the merge (step 9).
Step 1 — Review the working tree
git status
git diff
Read the full diff. Understand every changed file before touching git. Do not start staging until you know exactly how the commits will be split.
Step 2 — Create the branch
Branch off of current main. Always use a prefix:
What ships with it
1 file 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.
- 1mo ago First seen · 352 lines · 72 tokens per session scan A cf985678bd2d
ship-changes is a skill published in the GitHub repository Cratis/VerticalSlices (2 stars, last pushed 1mo ago), licensed MIT. It adds 72 tokens to every session and 3,770 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
cratis-governed-release-methodology
Choose the assurance tier a release needs, gather the evidence that tier requires, label release intent by semantic-version impact, and structure a canary-before-promotion path with an explicit recovery disposition. Use when planning a release or judging whether checksums, provenance, an SBOM, or a claim of support…
cratis-release-notes
Draft or review developer-facing release notes and migration guidance from verified changes for a specific Cratis product version. Use for GitHub releases, upgrade guides, and release announcements; not for deciding release authority, publishing, or selecting a version label.
cratis-engineering-consolidation-guard
Prove that a consolidation or re-land dropped nothing - compare the result with every original branch or commit, detect lost files that commit pairing cannot see, treat a merge resolution that reverts a file as a finding, ratchet the specification count, and enforce it as a required CI check. Use when folding several…
cratis-screenplay-command-surface
Write the write side of a Cratis Screenplay .play model — the command block and its identifier, reads, validate, authorize, produces, handler and concurrency clauses, plus event, constraint, policy, persona, concept, type and seed. Use when declaring or changing a command, an event shape, a validation or authorization…
cratis-screenplay-read-surface
Declare the read side of a Cratis Screenplay .play model — the readmodel shape, query with by/filter/scoped to/observable/performer, and screen at its three levels of detail, including the data/action contract that later component bindings, templates, forms, toolbars, routes and outlets consume. Use when adding or…
cratis-screenplay-ui-composition
Compose the user interface of a Cratis Screenplay .play model — layout and responsive arrangement, screen/dialog templates, component data-context bindings, exposed template configuration, command forms and columns, navigation contributions, application navigation regions (topbar/sidebar, module…