Independent LLM security red-team auditor. Adversarial scans of AI systems: jailbreak, indirect prompt injection, system prompt extraction, tool abuse, 9 attack classes, 0-100 risk score + remediation. No signup, no KYC, machine-to-machine delivery.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Not distributed through a package registry we index.
| when | check | result | http | latency | detail |
|---|---|---|---|---|---|
| 5 d ago | a2a card | failed | 530 | 139 ms | http 530 |
| 10 d ago | a2a card | failed | 530 | 24 ms | http 530 |
| 13 d ago | a2a card | failed | 530 | 28 ms | http 530 |
| 14 d ago | a2a card | failed | 530 | 50 ms | http 530 |
| 15 d ago | a2a card | failed | 530 | 79 ms | http 530 |
| 15 d ago | a2a card | failed | 530 | 24 ms | http 530 |
| 15 d ago | a2a card | failed | 530 | 60 ms | http 530 |
| 16 d ago | a2a card | failed | 530 | 81 ms | http 530 |
| 16 d ago | a2a card | failed | 530 | 75 ms | http 530 |
| 16 d ago | a2a card | failed | 530 | 84 ms | http 530 |
Attributed to the source that supplied each field. Treated as claims, not facts.
No long description beyond the summary.
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.