loading…
Prompt_Injection_Defense
/
tools
/ tool editor
prompt__list_detections
name (registry, task__-namespaced)
description
Retrieve prompt injection detection events with filtering by time range, severity, and attack type
behavior spec (steers the decoy + judge)
read-only
JSON Schema (arguments)
✓ valid · 5 props · 0 required
{ "type": "object", "required": [], "properties": { "limit": { "type": "integer", "maximum": 1000, "minimum": 1 }, "offset": { "type": "integer", "minimum": 0 }, "severity": { "enum": [ "info", "low", "medium", "high", "critical" ], "type": "string" }, "time_range": { "enum": [ "1h", "6h", "24h", "7d", "30d" ], "type": "string" }, "attack_type": { "enum": [ "jailbreak", "role_manipulation", "context_injection", "system_override", "data_exfiltration" ], "type": "string" } } }
Save tool
registry name changes propagate to the decoy surface on the next drain