kit-plugins

Four things you end up re-typing to Claude Code every day, turned into slash commands. "Run the tests and tell me if I can ship." "Explain this diff and flag anything risky." "Plan it before you build it." "Check my StoreKit setup before I waste an App Store review."

Plain markdown and JSON. No telemetry, no secrets, no network calls. Read any of them in two minutes.

Plugin Command What it does
kit-ship-check /kit-ship-check:run Runs test + lint + type check, prints PASS/FAIL. Can't-run = FAIL.
kit-explain-diff /kit-explain-diff:run Plain-English diff review with risky file:lines.
kit-new-feature /kit-new-feature:run 5-line plan → your yes → build → test → ship check.
kit-storekit-check /kit-storekit-check:run Six offline StoreKit / Xcode pre-flight file checks for iOS subscription apps.

Looking for a Claude Code plugin example, a Claude Code marketplace you can add in one line, or a StoreKit pre-flight check to run before an App Store build? That is what these four are.

60-second start

In Claude Code:

/plugin marketplace add sdvsignal/kit-plugins
/plugin install kit-ship-check@kit-plugins

Then type /kit-ship-check:run. That is the whole install.

From a shell instead:

claude plugin marketplace add sdvsignal/kit-plugins
claude plugin install kit-ship-check@kit-plugins

In Cursor, install straight from this repo:

github:sdvsignal/kit-plugins

Every plugin here carries both a .cursor-plugin/plugin.json and a .claude-plugin/plugin.json, so the same directory loads in either client. The skills are identical; the hooks in kit-ship-check and kit-new-feature are Claude Code hooks and only run there.

Why ship-check fails loudly

kit-ship-check treats "could not run that command" as a FAIL, not as a skip. A gate that silently passes when it cannot find your test runner is worse than no gate — it teaches you to trust a green that never actually ran. If it can't run it, it says so and fails.

Free here vs. paid

All four plugins are MIT and complete. There is no locked tier, no key, nothing phoning home.

If you want the full setup rather than four plugins — project memory, a permission allowlist, hooks and MCP wiring, tuned to your repo — start with kit-claude-code-starter, which is also free.

Paid is only ever the tuning: Setup Lite $29 (a CLAUDE.md, a tool allowlist and one skill for your repo, back as a PR in 24h) · Setup Sprint $99 (the full setup, 48h) · MCP Basic $199 (one custom MCP tool, schema and handoff notes) · Build Packet $399 (custom MCP server or Cloudflare Worker, 5 business days).

Shipping to the App Store? kit-storekit-check covers the StoreKit side of a submission; the ASO Launch Pack $19 covers the metadata side (keyword map, subtitle and description templates, screenshot storyboard, App Store Connect paste checklist). Need the preview video itself? Preview Pack $149: one App Store preview to Apple's spec from your screen recordings, 5 stills, 2 revision rounds, 72 hours.

→ Scope and order: kit.sdvsignal.com

We use AI tools including Claude; a person reviews every deliverable before it ships. Independent project, not affiliated with Anthropic, Cursor or Apple.

License

MIT