target-agent
2026-07-11 12:00
| Severity | Attack Name / ID | OWASP Category | Source | Status |
|---|
Timeline of continuous monitoring checks (`crucible watch`).
| Timestamp | Status | Score Delta | New Regressions |
|---|
High Risk (Art. 6 & 15)
Scans evaluate robust output alignment, memory poisoning thresholds, and parameter validations required for Article 15 compliance.
0 Triggered
Adversarial Machine Learning techniques identified (e.g. Indirect Prompt Injection AML.T0054, LLM Jailbreak AML.T0051).
MEASURE & MANAGE
Enables quantitative AI risk measurement (MEASURE 2.5) and policy boundary actions (MANAGE 2.2).