AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
🌱 Contract-first environment configuration with typed schemas
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
Contract-first environment configuration with typed schemas — runtime-agnostic core (Node, Bun, Deno, browser, Workers, Edge) with opt-in adapters for Node, Vite, and React (SSR).
This is the monorepo. Three things live here:
The whole thing is orchestrated with mise. Get the runtimes, install, run the e2e suites:
Run mise tasks for the full list.
For everything else — the per-task reference, the per-example layout, how to iterate on the library while seeing changes in the examples, troubleshooting — see DEVELOPMENT.md.查看完整 README ↗This repository holds env: the @variablelab/env library, its docs site and its examples.
pnpm add @variablelab/env
Usage and the entry points are in the library's README, and everything else is in the docs.
package/: the @variablelab/env library, published to npm.docsite/: the docs at lab.variable.land/env/docs, built with Fumadocs and served by Cloudflare Workers.examples/: runtime-isolated apps for Node, Bun, Deno, Workers, Edge, a Vite SPA and SSR. Each one installs the packed library, as an outside consumer would.The contributing guide covers issues, pull requests and releases, and the development guide covers setting up the repo and running its checks.
The code is MIT. The Sign in package/brand/ is not: its use is set by BRAND-USE.md.