By name: kmp-production-readiness description: Audit Kotlin Multiplatform production readiness with explicit scorecards, release blockers, risk ownership, validation commands, and deferred checks across architecture, build, testing, interop, security, performance, and publishing.
KMP Production Readiness
Bundled commands use $PLUGIN_ROOT ($env:PLUGIN_ROOT in PowerShell; same path suffix) for the plugin root. Set it once: use the host's plugin-root variable when defined (Claude Code: PLUGIN_ROOT="$CLAUDE_PLUGIN_ROOT"), otherwise the absolute path of this plugin's root directory.
Use this skill for full KMP production readiness audits, release-blocker triage, readiness scorecards, executive summaries, and multi-area risk reviews.
Flow
- Read project structure, Gradle files, version catalog,
gradle.properties, CI files, release scripts, and platform app shells. - Run:
python3 "$PLUGIN_ROOT/scripts/kmp_inspector.py" --root <project-root> --json --fail-on none - Classify the project:
- app-only shared code
- internal mobile shared library
- public KMP library
- shared Compose UI app
- mixed native UI plus shared logic
- monorepo with multiple KMP modules
- Map evidence into readiness areas:
- project structure
- build governance
- testing quality
- iOS/native interop
- security/privacy
- performance/observability
- publishing/release
Verdict Rules
ready: no release-blocking gaps in the area, validation commands exist, and residual risk is explicit.watch: shippable with known risk, owner, and follow-up command.blocked: missing validation, unsafe configuration, public API/publishing uncertainty, secret/security risk, or platform integration failure.
Do not average away a blocker. A single blocked area can block the overall verdict.
Output
Lead with:
- overall verdict
- blocked areas and exact blockers
- readiness scorecard
- validation commands already run
- validation commands still required
- release owner decisions needed
- deferred checks and why they are acceptable or not acceptable