🌱 kiwa v0.5 — polyglot test toolchain · Coverage + Mutation gates · 8 component adapters

[cardene777]

🌱 kiwa v0.5 — polyglot test toolchain · Coverage + Mutation gates · 8 component adapters

A scattered test stack is the default. Contract tests live in Foundry, unit + API in Vitest, e2e in Playwright, components in Testing Library, a11y in axe-core, visual in pixelmatch, Python services in pytest. Every runner has its own conventions, fixtures, and gates — and no single source of truth ties them together.

kiwa v0.5 is the cut where I finally feel the toolchain solves that. One Layer 1 spec → every test layer your stack actually needs, across TypeScript, Python, and Solidity.

1. Six test surfaces, one spec

Surface	Stacks
Contract	Foundry / Hardhat (Solidity)
API integration	msw / supertest / Vitest
Component	@kiwa-test/ui ships 8 adapters (see § 2)
E2E	Playwright + anvil + viem + EIP-6963 + ERC-4337
A11y + Visual	axe-core / pixelmatch
Data + CLI + Observability	queue / cron / shell IO / flaky detection

dApps and smart contracts are first-class — but so is everything else.

2. @kiwa-test/ui covers 8 component surfaces from one package

Framework	Helper
React	setupComponentEnv
Vue 3	setupVueComponentEnv
Svelte	setupSvelteComponentEnv
SolidJS	setupSolidComponentEnv
Lit (Web Components)	setupLitComponentEnv
Qwik (resumable)	setupQwikComponentEnv
Angular	setupAngularComponentEnv
Browser (real Chromium)	setupBrowserComponentEnv

All share the same mode (render / interaction / snapshot) + stop() contract. Optional peer deps mean you only install what you actually use.

3. Polyglot from day one

Language	Packages	Notes
TypeScript	11 npm packages	Primary surface — every adapter
Python	1 PyPI package (kiwa-test-py)	@kiwa-test/spec Python port + requests / httpx adapter
Solidity	Foundry / Hardhat bridges	/kiwa-forge + /kiwa-hardhat from the same Layer 1 spec

Rust / Go on the roadmap. The point: spec once, generate test layers across the languages your stack actually uses.

4. Release-time gates that actually fire

scripts/check-coverage-gates.mjs + scripts/check-mutation-gates.mjs run inside .github/workflows/release.yml and block publish if any package regresses below its threshold.

Coverage gate

Lines / Statements / Functions ≥ 90 %, Branches ≥ 80 %, across all 11 packages.

Mutation gate — Mutation Score Indicator (MSI)

Package	MSI	Threshold
@kiwa-test/api	96.06 %	90
@kiwa-test/a11y	93.62 %	90
@kiwa-test/ui	91.76 %	80
@kiwa-test/cli-test	89.69 %	80
@kiwa-test/data	86.93 %	80
@kiwa-test/spec	85.51 %	80
@kiwa-test/core	85.09 %	80
@kiwa-test/cli	84.44 %	80
@kiwa-test/e2e	84.21 %	80
@kiwa-test/observability	84.12 %	80
@kiwa-test/visual	83.02 %	80

Per-package thresholds are intentional — pure-logic packages enforce 90; thin wrappers around third-party libs hold at 80 to avoid getting trapped by equivalent mutants.

Try it

npm install @kiwa-test/core
# or for Python services
pip install kiwa-test-py
# scaffold everything
pnpm dlx @kiwa-test/cli init

• 📖 README · https://github.com/cardene777/kiwa
• 📦 npm · https://www.npmjs.com/package/@kiwa-test/core
• 🐍 PyPI · https://pypi.org/project/kiwa-test-py/
• 🎬 80s overview · assets/kiwa-promo-en.mp4 (see README inline)

How can you help?

• Try it on a stack with mixed test surfaces (dApp + API + a11y, or Python service + REST contract, or a multi-framework UI). File an Issue / Discussion if anything breaks or feels awkward.
• Adapter contributions welcome — @kiwa-test/ui is purpose-built for new framework adapters (the SolidJS adapter took 1 PR and 100 LOC).
• Polyglot bridges are the most strategic contribution surface. Rust / Go ports of @kiwa-test/spec are on the roadmap — let's talk in this thread if you want to drive one.
• Ideas for new test surfaces (gRPC mock / Pact contract test / Storybook visual diff) — discuss in an Idea Discussion.

Thanks for reading 🌱

— @cardene777