User repository on GitHub
Mock Jev API decisions with MSW for deterministic tests without real API calls or credits.
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Repos & SDKs
Jev Msw
Contract-aware MSW handlers for the official @typesafe-ai/sdk: mock Jev choices, Nouls and scores in tests without API calls, credentials, latency or credits, while the real SDK parsing and retry logic still runs.
- Stars
- 2
- Language
- TypeScript
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- royalpinto007
- Last verified
- Oct 2026
Works forJev for Evaluation
Used inJev for Software & DevTools
- Capabilities
- Open source
Highlights
- Provides jev.choice, jev.noul, jev.score and jev.respond handlers for the real @typesafe-ai/sdk, so parsing, error, timeout and retry logic still run.
- Missing probability maps, score legends, score probabilities, model and zero-token usage are derived from labels in the incoming question.
- jev.mock supports dynamic decisions with deep-partial or predicate matchers for state, questions and model, plus delay and attempt-aware answers.
- Failure helpers cover serverError, rateLimited, error(422), connectionError, timeout and malformed; sequence() replays a chain and sticks on the end.
- jev.history exposes requests and lastRequest; development setup needs only jev-msw and msw as dependencies.
Quickstart
ts
import { TypeSafeClient } from "@typesafe-ai/sdk";
import { jev } from "jev-msw";
const client = new TypeSafeClient({ apiKey: "test-key" });
server.use(jev.choice("category", { choice: "billing", confidence: 0.96 }));Watch out
Copy reviewed against the truncated repo README on 2026-10-15; the excerpt cuts off inside the request-inspection section, so one custom matcher export is only partially visible. The README positions the handlers as supporting the official SDK's documented retry surface rather than every SDK version.
Questions about this entry
Short answers drawn from the linked project or page.
A testing library that provides contract-aware MSW handlers for the official @typesafe-ai/sdk. It makes Jev-driven tests deterministic without API calls, credentials, latency or credits, while the application still runs the official SDK's real parsing and error logic.
npm install -D jev-msw msw, then set up an MSW server and use jev.choice, jev.noul, jev.score, jev.respond or jev.mock handlers in your tests. jev.reset() between tests and server.resetHandlers() keep state clean.
No. Your code still calls TypeSafeClient from @typesafe-ai/sdk, and its real parsing, error, timeout and retry logic runs; only the network responses are mocked by MSW handlers.
Choice probabilities, score legends, score probabilities, model and zero-token usage are completed from the incoming request unless you provide them explicitly; the missing probability map is derived from the actual labels in the question.
Yes. jev.mock takes a function of request and attempt and supports matchers for state (deep-partial), questions and model (predicates allowed), plus delay; the README example picks priority or standard based on state.vip and varies confidence per attempt.
Handlers include jev.serverError, jev.rateLimited with retryAfterMs, jev.error(422), jev.connectionError, jev.timeout and jev.malformed. jev.sequence replays handlers in order and keeps returning its final handler, so the SDK's 408, 429, 5xx and timeout retries are testable.
jev.history.requests lists captured requests and jev.history.lastRequest exposes the most recent one, so you can assert on the state or question your client actually sent.
More Jev resources like this
Offline Jev token counting and request accounting for Node.js and Next.js: an MIT TypeScript port of jevtok that matches the reconstructed encoder for jev-1.13-20260917 and ships its vocabulary inside the package — no keys, network calls or dependencies.
Jev Repos & SDKs#community#catalog-import#logicrw
Community
aarzhaev/jevtok-tsTypeScript
aarzhaev/jevtok-tsTypeScriptLangChain is the agent engineering platform: an MIT-licensed Python framework from the LangChain team for building agents and LLM applications by chaining interoperable components with third-party model, tool, and vector-store integrations.
Jev Repos & SDKs#community#catalog-import#cobanov
Community
langchain-ai/langchainPython
langchain-ai/langchainPythonComposio is an MIT-licensed SDK monorepo giving AI agents 1,000+ pre-authenticated toolkits with per-user sessions, authentication, triggers, and a sandbox, and it ships TypeScript and Python adapters including one for TypeSafe (Jev).
Jev Repos & SDKs#community#catalog-import#cobanov
Community
composiohq/composioTypeScript
composiohq/composioTypeScriptThe AI Toolkit for TypeScript: Vercel's free, open-source AI SDK for building AI-powered applications and agents, with GitHub topics covering OpenAI, Anthropic, Gemini, React, Svelte, Vue, and Next.js.
Jev Repos & SDKs#community#catalog-import#logicrw
Community
vercel/aiTypeScript
vercel/aiTypeScriptLangChain.js is the TypeScript edition of LangChain, a framework for LLM-powered applications that runs on Node.js, Cloudflare Workers, Vercel and Next.js, Supabase Edge Functions, browsers, Deno, and Bun.
Jev Repos & SDKs#community#catalog-import#cobanov
Communitylangchain-ai/langchainjsTypeScript
langchain-ai/langchainjsTypeScriptReqLLM is an Apache-2.0 Elixir library built on Req and Finch that standardizes LLM requests and responses across providers, exposing 1,205 models across 21 integrations behind one unified API.
Jev Repos & SDKs#community#catalog-import#logicrw
Community
agentjido/req_llmElixir
agentjido/req_llmElixirFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.