crucible_adversary
Adversarial testing and robustness framework for AI models with 25 attacks (character/word/semantic perturbations, prompt injection, jailbreak, extraction, inversion), defenses (detection/filtering/sanitization), certified robustness metrics, and attack composition.
View on Github
Activity
- Latest release
- 8mo ago
- Total releases
- 4
- Cadence
- ~32 days
- Last 12 months
- 4
Reach
- Stars
- —
Details
- License
- MIT
- First release
- Oct 11, 2025
Releases
| Version | Released | |
|---|---|---|
0.4.0
minor
| ||
0.3.0
minor
| ||
0.2.0
minor
| ||
0.1.0
initial
|