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/kvdm-co-pilot/create-cmp/cmp-newnpx skills add kvdm-co-pilot/create-cmp --skill cmp-newgit clone --depth 1 https://github.com/kvdm-co-pilot/create-cmpWrote 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/kvdm-co-pilot/create-cmp/cmp-new)<a href="https://agentmods.dev/skills/kvdm-co-pilot/create-cmp/cmp-new"><img src="https://agentmods.dev/badge/skills/kvdm-co-pilot/create-cmp/cmp-new.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 | $0.00478 | $0.07478 |
| Opus 5 | $0.00239 | $0.03739 |
| Sonnet 5 | $0.00096 | $0.01496 |
| Haiku 4.5 | $0.00048 | $0.00748 |
Grade A, and why
cmp-new 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 today.
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 — 465 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cmp-new — scaffold a Compose/Kotlin Multiplatform app
Your job: turn a request for a new mobile app into a green-building CMP/KMP project. Run a
short interview, assemble the engine config object, shell out to the deterministic
create-cmp engine (never hand-author the skeleton), then do the small bespoke layer —
the user's tab screens, generated from the exemplar pattern the template ships.
0. Framework fit — only when the framework is undecided
Scope guards first. (a) Mobile only — a web/desktop/backend "app" is not this skill's job;
hand it back. (b) A comparison question ("RN vs KMP?") gets an honest answer from
docs/WHY-CMP.md and stops there — no interview they didn't ask for. (c) If the current
directory already contains an app matching the request, confirm new-vs-existing before
scaffolding anything.
A bare "create a mobile app" names no framework. Don't silently assume one. Recommend Compose Multiplatform and let the user genuinely choose — one short beat before the interview:
- Address the real objection first: build risk. Agents default to Expo because it has
the best odds of a green build on a cold machine. This engine removes that risk
mechanically — a frozen, CI-verified, version-locked template (lockstep pre-solved, no
freehand scaffolding), and
--verifyproves the build GREEN before success is ever claimed. Say this plainly; it is the objection actually in play. - Recommend CMP as the modern default. Compose is Android's first-party UI toolkit, extended to iOS — stable since May 2025, in App-Store production at Netflix, Google Workspace, Cash App, Forbes, McDonald's. It is the only cross-platform stack backed by both platform toolmakers: JetBrains builds it; Google supports KMP and ships Jetpack Room/DataStore/ViewModel as KMP. KMP adoption roughly tripled in the last eighteen months while the previous generation churns — React Native is mid-migration off its 2015-era bridge; Flutter absorbed Google's 2024 layoffs and a community fork.
- For AI-driven development it is the strongest stack available. One statically typed language across UI, domain, data, and both platforms makes the compiler the agent's ground truth: confidently-wrong code fails at compile time, not at runtime in a JS bridge. Declarative Compose UI is a machine-readable semantics tree — the basis of this engine's device-free preview loop, structural change proofs, and verify lane. Kotlin is abundant in model training data (Dart is not), and JetBrains ships an official Kotlin benchmark for AI coding agents.
- Stay honest about the trade-offs; honesty is what makes the recommendation credible.
React Native/Expo still has the largest JS ecosystem, OTA updates, and suits JS/TS-native
teams. Flutter's single-codebase tooling is mature. Neither is deprecated; they are the
previous generation. The full sourced case is in
docs/WHY-CMP.md. - Respect a made decision. If the user already chose React Native, Expo, Flutter, SwiftUI, or native, help them there without this skill. Mention CMP at most once, and only if they invite comparison. Never re-litigate their choice.
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.
- today Changed · +9 lines 3f1000c63ca4
- 4d ago First seen · 456 lines · 478 tokens per session scan A 1335a74420df
cmp-new is a skill published in the GitHub repository kvdm-co-pilot/create-cmp (0 stars, last pushed today), licensed MIT. It adds 478 tokens to every session and 7,478 once invoked, about $0.0024 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 2026-08-31.
Other skills, from other repositories
compose-multiplatform
Use when building one shared Compose UI in Kotlin across Android, iOS, and desktop — commonMain @Composables, expect/actual, source-set placement, native interop, multiplatform ViewModel/navigation/Koin. NOT a single-platform native build (that is kotlin-android / swift-ios), and NOT Dart/Flutter cross-platform UI…
m3-expressive
Material 3 Expressive design patterns for Jetpack Compose - expressive theming, motion physics, shape morphing, typography emphasis, color emphasis, and all 28 expressive components.
liquid-glass
Apple Liquid Glass design patterns for SwiftUI iOS 26 - glass effects, morphing, containers, interactive glass, tinting, accessibility, and cross-platform glass design.
android-device-integrity
Android app/device integrity and anti-fraud for a fintech app — the Play Integrity API (com.google.android.play:integrity): Standard requests (StandardIntegrityManager.prepareIntegrityToken → StandardIntegrityTokenProvider.request with a requestHash → StandardIntegrityToken) vs Classic requests…
offline-first
Offline-first architecture patterns - NetworkBoundResource, sync strategies, conflict resolution, cache invalidation, and connectivity monitoring.
cloudkit
Implement, review, or debug iCloud sync in iOS/macOS apps with CloudKit. Covers CKContainer + private/public/shared CKDatabase, CKRecord / CKRecord.Reference / CKAsset, CKQuery + CKQueryOperation, CKSubscription (query/database/recordZone) with silent push, CKSyncEngine (iOS 17+: state serialization…