AI & Agents
HackMyClaw
Last updated 2026-06-28 · benchmark measured 2026-08-09 — deterministic & reproducible
A now-closed prompt injection challenge where participants tried to make an AI leak secrets via email.
Is HackMyClaw production-ready?
Legit.Show scores HackMyClaw 73 out of 100 — the simple average of its 7 measured frames. Legit.Show ran its deterministic 7-Frame production-readiness benchmark on HackMyClaw (public-surface assessment), measured from the public surface with no LLM in the scoring path. Its strongest frame is Discoverability; its weakest is Privacy. Every frame it averages is published with its evidence on the Legit.Show listing.
The 7 Frames
- Performance — 93/100
- Accessibility — 96/100
- Security — 45/100
- Privacy — 25/100
- Reliability — 60/100
- Standards — 92/100
- Discoverability — 100/100
What we measured
- Security headers present: X-Content-Type-Options, Referrer-Policy.
- No Content-Security-Policy and no HSTS.
- Served over HTTPS with a valid certificate.
- Real Lighthouse performance run — 214 ms to first byte.
- 1 of 2 sampled routes reachable.
- No privacy policy found.
- Sets cookies / loads scripts with no consent prompt.
- Discoverable: structured data, sitemap, OpenGraph image, canonical URL.
Who built it
Fernando Irarrázaval (@Cucho)
Who it's for
Security researchers · AI safety testers · Prompt injection enthusiasts · Developers · Hackers
Pricing
$1,400 total prize pool (now closed)
Visit HackMyClaw → · Alternatives to HackMyClaw → · How this was measured →