yanglei070-ux/bili-hardcore-jev
User repository on GitHub
用 JEV（TypeSafe System One 判断模型）自动完成 B 站「硬核会员试炼」的单文件 Python 脚本：手机只扫码登录一次，100 题在电脑上自动答完，零第三方依赖。
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Tools & CLIs
Bili Hardcore Jev
A single-file Python script that signs into Bilibili's hardcore member quiz after one QR scan, sends each multiple-choice question to Jev for a pick and a confidence, and submits a full round unattended.
- Stars
- 2
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- yanglei070-ux
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- Python
Highlights
- The quiz only opens in Bilibili's phone app, so the script presents itself as the Android client and calls the fetch, submit and score endpoints directly.
- One scan on the phone is the only manual step, and the login state is stored locally and expires after seven days.
- Each question prints its pick, the confidence and a running correct count, then the report gives the total and the per-section scores.
- It decodes real API error codes: 41103 already answered, 41104 answered too fast, 41105 finished, 41099 out of attempts.
Watch out
Bilibili's own rules state that using abnormal technical means during testing can get an account permanently barred from the challenge, and the README leaves that judgement to you. The endpoints are reverse-engineered, so a server-side change breaks the script, and the appsign layer is ported from another project. It needs a level 6 account, three attempts a day, one of which each run consumes, and a Jev key; membership itself lasts 365 days. Credentials are written to ~/.bili-hardcore/auth.json and the key to a file whose first line is the key, and neither should be shared. A captcha still has to be solved by hand once per round, since submission only counts after it passes. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
That risk is stated by the project rather than dismissed. The README quotes the rule that anyone found using abnormal technical means during testing may be permanently banned from the challenge, and says using the script is your call. Mechanically it identifies itself as the Android app, which is also why a change on Bilibili's side is the most likely failure mode rather than a subtle detection.
Only the pick. Jev is given the question and its options and returns which to choose plus a confidence number, with no prose, which is why the README calls the call cost across a hundred questions negligible. The script does everything else: fetching, the captcha, submitting, counting, scoring, and translating error codes into retries, waits or wrap-up.
One, and the script makes you confirm before it starts because the attempt only counts after you press enter. You also pick which sections to answer, one to three of them, and pass a captcha first, since Bilibili only counts the round once the captcha is submitted. If a run is interrupted, rerunning detects the unfinished round and offers to continue without spending another attempt, or to abandon it and start over.
The login state goes to ~/.bili-hardcore/auth.json and expires after seven days, with --logout deleting it. The Jev key goes in a separate text file whose first line is the key, placed next to the script as jev.txt and gitignored, or passed with --key-file or JEV_KEY_FILE. The README says not to share either file, and run output goes to ~/.bili-hardcore/run.log so failures are shareable without sharing credentials.
Pick one to three of eight sections, answer up to 100 questions within 120 minutes, and pass by getting 60 or more correct. Each run consumes one of three daily attempts, and passing grants hardcore membership for 365 days. The script prints the total and per-section scores at the end.
Four things: scan one QR code with the phone app to log in, press enter to start the round, choose which sections to answer, and type one image captcha. Bilibili only counts the round once the captcha is submitted, so nothing runs unattended until that step passes.
The self-test checks the key, QR generation and a tiny Jev call without touching the account; status shows login state, whether hardcore membership is active and how many attempts remain today; result queries the score; and logout deletes the locally saved login credentials.
More Jev resources like this
34GitHub stars
Pure-HTTP automation for the TypeSafe register, magic-link login, onboarding, and API-key creation flow, with pluggable temp-mail backends, concurrent batch runs, and key verification exports.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
2951461586/Jev-Register-ToolPython
2951461586/Jev-Register-ToolPython36GitHub stars
Python toolkit plus 0.8B multimodal model that takes a message, document, or image and returns calibrated choice, yes/no, and score answers in one forward pass for local decision workloads.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
PsiACE/dohnutsPython
PsiACE/dohnutsPython39GitHub stars
Video tool where Jev scores subtitle segments and code assembles a short highlights cut plus a de-watered full cut with timestamped summaries.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
cclank/jevclipPython
cclank/jevclipPython30GitHub stars
Forkable weekday radar that asks Jev one yes-or-no question per interest over each new arXiv announcement and publishes a page plus RSS feed.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
Eliot5566/JEV-Paper-RadarPython
Eliot5566/JEV-Paper-RadarPython38GitHub stars
Async supervisory layer for Hermes Agent that locks Definitions of Done, proves mechanical criteria locally, and calls Jev only for unresolved semantic checks under a token budget.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
keeltrace/hermes-nervePython
keeltrace/hermes-nervePython15GitHub stars
Two-tier decision engine pairing a 22.6M-parameter MiniLM encoder on the Apple Neural Engine with the on-device Apple Foundation Model as fallback for uncertain inputs.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
kerryrm/systemANEPython
kerryrm/systemANEPythonFrom the Jev guides
Original write-ups on what this entry is for.
The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.
Jev Tools & CLIs#gateway#vercel#cloudflare
Read article
Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.
Jev Tools & CLIs#open-models#local#onnx
Read article
Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.
Jev Tools & CLIs#classification#email#security
Read article
From the Jev community
Posts from builders shipping with Jev right now.