Survival LLM
A free AI assistant and survival library (1.5 million offline articles) that runs on your own computer, with no internet. Shows the source of every answer. Mac and Windows.
Answer 10 short questions and see how you tend to think, feel and work with people. It takes about 2 minutes.
OtherTypeScript · JavaScriptGitHub @guventechOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Who Am I? — honest 10-question Big Five sketch (prototype)
Translation pending
Taken from the repository README
Loading the full README.
Read the full README ↗An English-only, 10-question Big Five prototype. Model parameters are not calibrated yet.
The flow is simple: answer questions, see results, copy them if wanted. New answers and results are never saved on the server. They stay in the current browser session, including across refreshes. Closing the tab normally clears them.
pnpm install
pnpm dev
pnpm lint
pnpm typecheck
pnpm test
pnpm e2e
pnpm benchmark
Set PLAYWRIGHT_CHANNEL=chrome to test using an installed Chrome. The browser tests cover desktop and mobile, refresh, keyboard navigation, accessibility, and zero storage requests even when an old consent choice exists.
Vercel builds with pnpm build, as specified in vercel.json. It no longer deploys Convex as part of an app build. Scoring happens in a short-lived browser worker, so quiz traffic does not cause server scoring or database writes.
This prototype requires explicit ALLOW_UNCALIBRATED=true and ALLOW_DRAFT_CONTENT=true at build time. Keep its early-version notice and noindex until real calibration and content review are complete. The deeper-analysis code remains available but is off in production by default.
Private briefs, secrets, and test artifacts are excluded from upload by .vercelignore. Branding and local favicon assets refer to bircans.com.
The older version did save some results. Existing records remain untouched, with their original retention job. Privacy offers a collapsed deletion form for people who kept their code. Only that legacy deletion endpoint needs the existing server-only CONVEX_URL, CONVEX_SERVER_KEY, and SESSION_CODE_PEPPER. Keep the pepper stable so existing deletion codes continue to work.
New save and feedback endpoints return HTTP 410, even if old credentials or consent choices are present. The quiz works without Convex. Do not close the shared Convex account or change other projects.
For production browser verification, build with the prototype flags, set PLAYWRIGHT_PRODUCTION=true, and run pnpm e2e. This verifies the direct question-to-results flow and no storage requests.
See REPORT.md for history, NOTICE.md for IPIP attribution, and LICENSE.