Developer Tools
Codex plays NetHack
Last updated 2026-07-27 · benchmark measured 2026-08-09 — deterministic & reproducible
Live asciinema stream of OpenAI Codex autonomously playing NetHack.
Is Codex plays NetHack production-ready?
Legit.Show scores Codex plays NetHack 74 out of 100 — the simple average of its 7 measured frames. Legit.Show ran its deterministic 7-Frame production-readiness benchmark on Codex plays NetHack (public-surface assessment), measured from the public surface with no LLM in the scoring path. Its strongest frame is Reliability; its weakest is Discoverability. Every frame it averages is published with its evidence on the Legit.Show listing.
The 7 Frames
- Performance — 78/100
- Accessibility — 84/100
- Security — 85/100
- Privacy — 70/100
- Reliability — 100/100
- Standards — 89/100
- Discoverability — 15/100
What we measured
- Security headers present: CSP, HSTS, X-Frame-Options, X-Content-Type-Options, Referrer-Policy.
- Served over HTTPS with a valid certificate.
- Real Lighthouse performance run — 1445 ms to first byte.
- Returns a proper 404 for unknown routes.
- 3 of 3 sampled routes reachable.
- Has a reachable privacy policy.
- Sets cookies / loads scripts with no consent prompt.
Who it's for
AI researchers · Game developers · Terminal enthusiasts · AI experimentation
Visit Codex plays NetHack → · Alternatives to Codex plays NetHack → · How this was measured →