Script Kitty OS
- 15 followers
- United States of America
- https://scriptkittyos.com
- @ScriptKittyOS
- dev@scriptkittyos.com
Popular repositories Loading
-
HolyTrinity-Benchmark
HolyTrinity-Benchmark PublicAdversarial security benchmark for agent authorization: does a compromised agent's policy-violating proposal become an unauthorized external effect? 73 trials, nine families, an independent oracle,…
-
Repositories
Showing 2 of 2 repositories
- .github Public
- HolyTrinity-Benchmark Public
Adversarial security benchmark for agent authorization: does a compromised agent's policy-violating proposal become an unauthorized external effect? 73 trials, nine families, an independent oracle, per-mechanism ablation, confidence intervals. 0 unauthorized effects in 61 attack trials (95% CI [0.0%, 5.9%]). Reproduction is partial.
Top languages
Loading…
Most used topics
Loading…