User repository on GitHub
Describe your startup idea. Jev decides: kill it, fix it or ship it.
- Forks
- 36
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Cookbooks
Killmyidea
Describe a startup idea and Jev returns KILL IT, FIX IT, or SHIP IT from one request of ten scored questions, with goal-weighted scoring and a SQLite archive.
- Stars
- 262
- Language
- TypeScript
- Reading time
- 3 min
- Published by
- Community
- Author
- monteduro
- Last verified
- Sep 2026
About this project
Describe a startup idea. Jev decides: KILL IT, FIX IT or SHIP IT. No generative LLM. One request asks TypeSafe Jev 10 questions in parallel: 8 plain indie-hacker questions scored 0–4, plus the category and whether the idea is understandable. The key is only read by api/evaluate.ts and never reaches the browser.
Quoted from the project's README. Read the full README.
Works forJev for Scoring
Used inJev for Data & Analytics
Highlights
- One request asks 10 questions: eight indie-hacker questions scored 0 to 4 plus category and understandability.
- The weighted average times 25 maps to KILL below 50, FIX from 50 to 64, and SHIP at 65 or higher.
- A goal parameter changes the weights: money by default, open_source swaps in adoption, and fun adds appeal.
- Successful evaluations land in data/analytics.sqlite, and TYPESAFE_MOCK=1 returns deterministic keyless answers.
Quickstart
bash
npm install
cp .env.example .env
npm run devWatch out
Needs a TYPESAFE_API_KEY, and the SQLite archive wants a persistent disk that serverless hosts do not provide. The repo ships no license.
Questions about this entry
Short answers drawn from the linked project or page.
No generative LLM is involved. One request asks Jev 10 questions in parallel: eight indie-hacker dimensions scored 0 to 4, plus the category and whether the idea is understandable. Each score is multiplied by 25, combined as a weighted average, and mapped to thresholds you can read in the source.
Treat it as one input. The thresholds are fixed in code: below 50 is KILL, 50 to 64 is FIX, 65 or higher is SHIP, and if understandability falls below 0.3 the raw scores are kept for analysis while the UI asks for more detail instead of showing a verdict. There is no evaluation against real startup outcomes; the shipped benchmark uses synthetic cases.
Yes, by default. Successful evaluations are archived server-side in SQLite at data/analytics.sqlite with the idea, scores, latency and the full diagnostic JSON, deliberately excluding IP addresses, cookies and browser identifiers. There is an explicit opt-out checkbox, and private history stays in the browser's localStorage with no accounts.
Not durably. Vercel Functions have no durable filesystem, so SQLite needs a persistent deployment or a replaced archive adapter before you rely on the collected data. There is also TYPESAFE_MOCK=1, which returns deterministic fake answers for keyless local UI work and is ignored on Vercel.
Each question's Jev score of 0 to 4 is multiplied by 25 to give 0 to 100, then the eight are combined as a weighted average. KILL is below 50, FIX is 50-64 and SHIP is 65 or above.
The API takes money (default), open_source or fun. Make money weights Real problem and Money twice, Open source replaces Money with Adoption, and Just for fun replaces Real problem and Money with Immediate appeal and Fun.
Yes. Set TYPESAFE_MOCK=1 and the API returns deterministic fake answers labelled MOCK DATA in the raw-data panel, though mock mode is ignored on Vercel.
More Jev resources like this
A chat-shaped command bar where Jev picks the tool, arguments, confirmation, and reply type each turn, and code calls the MCP servers, so no LLM ever writes the reply text.
Jev Cookbooks#community#typescript#mcp
Community
w3cj/jev-chatTypeScript
w3cj/jev-chatTypeScriptA small demo app that fetches a URL, turns the page into clean Markdown, and asks Jev ten bounded Choice questions about what a first-time visitor is likely to understand.
Jev Cookbooks#community#typescript#demo
Cloudflare Worker and React app that sends pasted corporate prose through about 45 typed Jev questions, then emits rustc-style diagnostics for passive-aggression, urgency and information density.
Jev Cookbooks#community#typescript#cloudflare
Community
asfarsadewa/human-compilerTypeScript
asfarsadewa/human-compilerTypeScriptA multiplayer demo workflow builder for Jev built with Liveblocks: wire Jev and LLM steps together, call the workflow through a REST API, and preview test runs with shared state and cursors.
Jev Cookbooks#community#typescript#demo
Live Hyperliquid desk across five isolated wallets where each tick packages the book, tape, and position as state and Jev answers Choice questions for long/short, open/close/hold, and leverage; hold sends no order. A missing private key means a dry run.
Jev Cookbooks#community#typescript#trading
Community
aowang-ai/jev-tradeTypeScript
aowang-ai/jev-tradeTypeScriptA demo application that navigates a Neo4j graph one hop at a time: Jev gets the outgoing relationships as a Choice plus a goal Noul in one system_one call, and beam search follows the probabilities.
Jev Cookbooks#community#python#neo4j
Community
jexp/neo4jevJupyter Notebook
jexp/neo4jevJupyter NotebookFrom 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.