ship-it
Skill0 starsChanged 46 days agoMITView on GitHub
Take a green project to a released, deployed artifact by orchestrating the release skills you have installed to do ALL the reversible pre-flight perfectly (preflight gate, version bump, changelog, cross-platform build, checksums, installer verify, preview deploy, drafted release + store metadata) and then handing you the exact publish buttons. Use when "ship it", "cut a release", "prepare to launch", "tag a release", "deploy to production", "submit to the store", "publish".
- Commits and pushes
46 other projects use a version of this.
Adopted by
Most recently taken up by
- alexandrosandre/agentic-engineeringtheir own version, first seen Nov 8, 2026
- alexandrosandre/alice-issue-agenttheir own version, first seen Nov 8, 2026
- spencerldixon/skillstheir own version, first seen Nov 8, 2026
- unional/skillstheir own version, first seen Nov 6, 2026
- frappe/skillstheir own version, first seen Nov 1, 2026
- schalk-conradie/skillstheir own version, first seen Nov 1, 2026
Used with
Projects with it also use modern-go, loop-it and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
jidechaoSkillmodern-go
Modernize Go code by applying version-appropriate idioms and APIs (gofix-style transformations). Scans go.mod for the Go version, then transforms Go source files to use modern patterns—from Go 1.0 through 1.26+. Use when the user says "现代化","现代Go语言", "地道的", "idiomatic", "modernize", "modern-go", "update Go code", "gofix", or wants to upgrade Go idioms.
Also in 3 projects that use this
jidechaoSkillloop-it
Automated issue loop with checkpoint/resume: fetch open GitHub issues → dependency-aware topological sort → implement each issue end-to-end → review with /review-it → document with /note-it → ship with /ship-it → repeat. Persists state to .loop-state.json for crash recovery. Triggers on: loop-it, loop issues, auto implement, 批量实现, 循环实现, 实现所有issue, 恢复循环, resume loop.
Also in 3 projects that use this
SentrySkillcreate-pr
Create pull requests following Sentry conventions. Use when opening PRs, writing PR descriptions, or preparing changes for review. Follows Sentry's code review guidelines.
Also in 5 projects that use this
GitHubSkillrefactor
Surgical code refactoring to improve maintainability without changing behavior. Covers extracting functions, renaming variables, breaking down god functions, improving type safety, eliminating code smells, and applying design patterns. Less drastic than repo-rebuilder; use for gradual improvements.
Also in 4 projects that use this
GoogleSkillto-issues
Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.
Also in 3 projects that use this
Hugging FaceSkillsimplify
Refactor code after tests pass. The "Refactor" phase of Red-Green-Refactor.
Also in 3 projects that use this
- +2 more from jidechao
Other versions46 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 46
- vectorize-io/hindsight★ 60ksame name
- learntocloud/learn-to-cloud-app★ 1.1ksame name
- leonvanzyl/agentic-coding-starter-kit★ 447same name
- smallnest/goal-workflow★ 283same name
- frappe/skills★ 162same name
- DECRUX9812/hermes-plugin-gameforge★ 44same name
- awslabs/llm-evaluation-system★ 26same name
- LunkiBR/ship-it★ 18same name
- paultyng/skill-issue★ 8same name
- davidmeirlevy/obvious-skills★ 6same name
36 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository51 other files it gives its agents.
- advisor-councilSkill
- advisor-persona-buildSkill
- agent-exploit-pathsSkill
- agentic-loopsSkill
- ai-receptionist-tenant-onboardingSkill
- audit-lensSkill
- beadsmithSkill
- bug-sweepSkill
- bug-zeroSkill
- build-in-public-scope-guardSkill
- consulting-sales-playbookSkill
- deepenSkill
- deposit-first-shippingSkill
- design-intelSkill
- dueling-idea-forgeSkill
- forgeSkill
- hotspot-scoutSkill
- idea-forgeSkill
- incident-zeroSkill
- kickstartSkill
- local-llm-opsSkill
- memguardSkill
- memory-forgeSkill
- migrate-proofSkill
- no-slop-bug-hunterSkill
- optimize-proofSkill
- perf-maxSkill
- polish-passSkill
- problem-breakthroughSkill
- project-autopilotSkill
- qwen-bug-zeroSkill
- qwen-forgeSkill
- qwen-kickstartSkill
- qwen-perf-maxSkill
- qwen-project-autopilotSkill
- qwen-ship-itSkill
- qwen-ux-maximizerSkill
- readme-realSkill
- reality-checkSkill
- recallSkill
- security-sweepSkill
- session-mineSkill
- simplify-proofSkill
- skillifySkill
- stream-demo-prepSkill
- swarm-runSkill
- test-maxSkill
- trace-itSkill
- unslopSkill
- ux-maximizerSkill
- ux-scanSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 42 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 36
/test-max: before its green counts. This gate is only as strong as the tests behind it.
- Distinctive: line 39
Clean tree + rollback path named: before any outward step.
“Common” is measured: kept by at least 6 in 10 of the 42 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Ship It
The last-mile vertical. The quality siblings (bug-zero / perf-max / ux-maximizer) get a project to good; ship-it gets it out the door. It conducts your release skills to do every reversible step flawlessly, then stops at each outward button and hands it to the operator. Shipping is irreversible and outward-facing, so the button is always the operator's.
Core Insight: ship-it is the inverse of the quality verticals. They're mostly-auto with a few judgment calls; this is mostly ask-user, because publish / prod-deploy / store-submit / cargo-publish / push are forbidden-to-auto by the rails. Its value isn't pressing buttons; it's making the button a 30-second decision instead of an hour of prep. Everything reversible (version, changelog, build, checksums, preview deploy, drafted release) is auto; everything outward is a staged button with a rollback path. Like its siblings it has a green ratchet and a decision inbox; here the inbox is mostly "here are the buttons."
Preconditions / guards (the go/no-go gate)
- Green + actually delivers, and the green is TRUSTED. Tests green AND a reality-check says the build delivers its intent (don't ship broken, and don't ship something that doesn't do what it claims). If pre-ship hardening is warranted, run /bug-zero and/or security-sweep FIRST. And if the suite's green can't be trusted (weak assertions, flakes, line-coverage theater), run /test-max before its green counts. This gate is only as strong as the tests behind it.
- No secrets in the release. Scan artifacts/config; never ship a key.
- Clean tree + rollback path named before any outward step.
- A release that includes a migration ships under /migrate-proof. Its expand → migrate → contract gate wraps the release: expand steps may ride this one; contract (killing the old path) waits for the parity oracle + a rehearsed rollback, and every contract step is a button.
- The go is the operator's. ship-it prepares and recommends; it never decides to ship.
The Pipeline (staged: reversible work auto; outward steps = buttons)
Phase 0 PREFLIGHT Go/no-go gate: tests green AND trusted (test-max if not) ·
reality-check delivers · pre-ship hardening done if warranted ·
any migration gated by migrate-proof · no secrets · clean tree. Output a
one-line go/no-go recommendation (the GO is the operator's). [auto check]
Phase 1 VERSION+NOTES Bump version (semver), write release notes / changelog
(changelog-md-workmanship), clean conventional commit
(git-commit-craftsman). [auto, reversible]
Phase 2 BUILD Build release artifacts: cross-platform binaries via gh-actions
CI (fallback: dsr/rch if installed, else your normal CI or a
local build); checksums; installer (installer-workmanship)
and VERIFY the installer actually works. [auto, reversible]
Phase 3 STAGE Everything that surrounds the release but stops short of publish:
DRAFT the GH release (unpublished), draft store metadata, deploy
to a PREVIEW env (vercel preview / staging, never prod), prep
OG/share images. [auto, reversible]
Phase 4 THE BUTTONS Present the operator the ready package with the EXACT publish actions and
each one's rollback path: push the tag · publish the GH release ·
promote preview→production · submit to the store · cargo publish.
ship-it does NOT press these. On the operator's explicit per-action go, it
may run the low-risk reversible ones (push a tag); the truly
irreversible/outward ones (cargo publish, store submit, prod
deploy) stay manual or need per-action confirmation. [ask-user]
Phase 5 POST-SHIP After the operator ships: smoke the released artifact / deployed URL /
installed binary; record the version + rollback path; note any
platform still in review (decoupled: one platform's limbo never
blocks another). [auto verify]Button discipline (the one rule that matters)
Reversible & in-repo (version, changelog, build, checksums, preview deploy, drafted
release) → auto. Outward & irreversible (publish, prod deploy, store submit,
cargo/npm publish, git push of tags/releases, force-push) → never auto; surface
as a button with its exact command and rollback. When unsure which side a step is on,
treat it as a button. Ship small + platform-decoupled (the standing cadence): internal →
beta → prod, promote fast when the user base is tiny, and never let one store's review
freeze another platform's release.
Skills it orchestrates
Map + staged prompts in references/ROUTING.md. These release skills are not part of this pack: use each if installed; otherwise run the same step directly with git, gh, and your build tool. Spine: release-preparations (tests, version, cross-platform build, checksums, GH release, installer verify). Plus changelog-md-workmanship · git-commit-craftsman · gh-actions · dsr / rch (optional remote-build helpers: GH-Actions fallback when installed; otherwise local builds or your own CI) · installer-workmanship · rust-crates-publishing · gh-cli · vercel / wrangler / gcloud (deploy targets) · gh-og-share-images. Pre-ship gates: /bug-zero · security-sweep · reality-check · /test-max (test-trust) · /migrate-proof (when the release carries a migration).
Shared mechanisms (inherited from project-autopilot)
Green ratchet, decision inbox, capability-aware execution, as in project-autopilot (see its CONVERGENCE / SAFETY-RAILS / EXECUTION). Integration: project-autopilot's "Release/tag/publish" route points here; ship-it is the lifecycle vertical that closes build → ... → SHIP, alongside the three quality verticals.
THE EXACT PROMPT
/ship-it
Target: <project path>
Channel: <internal | beta | production | crates.io | app-store | npm | …>
Mode: solo | swarmAnti-Patterns
- Pressing an outward button (publish / prod deploy / submit / cargo publish / push) without the operator's explicit per-action go: the cardinal sin.
- Shipping a red build, or one a reality-check says doesn't deliver.
- Shipping with a secret in the artifact/config.
- Deploying straight to prod instead of a preview first.
- Big-bang hoarding instead of small numbered increments.
- Letting one platform's review limbo block another's release (keep them decoupled).
- Skipping the post-ship smoke (a published thing you didn't verify isn't shipped).
References
- ROUTING.md: phase → skill map, the button list, staged prompts
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

