loading…
LLM attack
/
tools
/ tool editor
llm__search_attacks
name (registry, task__-namespaced)
description
Search attack campaigns by target model, technique, or adversary with filtering and pagination
behavior spec (steers the decoy + judge)
read-only
JSON Schema (arguments)
✓ valid · 7 props · 1 required
{ "type": "object", "required": [ "query" ], "properties": { "limit": { "type": "integer", "default": 50 }, "query": { "type": "string", "description": "Search term for attack name, technique, or target" }, "offset": { "type": "integer", "default": 0 }, "status": { "enum": [ "active", "mitigated", "investigating", "false_positive" ], "type": "string" }, "severity": { "enum": [ "critical", "high", "medium", "low" ], "type": "string" }, "technique": { "enum": [ "prompt_injection", "jailbreak", "data_extraction", "model_inversion", "membership_inference", "backdoor", "adversarial_examples" ], "type": "string" }, "target_model": { "type": "string" } } }
Save tool
registry name changes propagate to the decoy surface on the next drain