Bug Hunt Bench

Bug Hunt Bench: 105 real bugs in two production repos, frontier coding models (GPT-6, Claude, Grok, Gemini, DeepSeek...) find and fix them in their own CLI, graded blind. Live leaderboard + every receipt.

Description excerpted from the original listing, which is linked below.

Open it Listing

Facts

Published: 25 Aug 2026 (18d ago)
Medium: web
Language: English (the page says so itself)
First seen: 12 Sep 2026, 15:12 UTC

Built with

Claude Code: confirmed
the agent's own trailer in the repository's commit history

What we can see

94 hygiene score

Certificate issued by YE1, 71 days remaining
Cookie attributes no cookies set
Credentials and payment does not ask for either
Strict transport security set
Served over HTTPS
Mail authentication neither published
No insecure resources
Redirects none
Security headers 5 of 5 set
Third-party origins 3
Transport security TLS 1.3

Observed 12 Sep 2026 by fetching the page as a browser would. This describes hygiene and transparency, not safety: a missing header does not make a site dangerous, and a complete set does not make it trustworthy.

Changes we have seen

12 Sep 2026 first indexed