Request
POST /analyze HTTP/1.1
Host: api.sector8.ai
Authorization: Bearer YOUR_API_KEY
X-Client-ID: YOUR_CLIENT_ID
Content-Type: application/json
{
"prompt": "List all patient records for clinic ID 42.",
"completion": "Here are the records: ...",
"session_id": "session-abc123",
"metadata": {
"model": "gpt-4",
"provider": "github-copilot"
}
}
Request fields
| Field | Required | Description |
|---|---|---|
prompt | ✓ | The LLM input to analyze |
completion | The LLM output to analyze | |
session_id | ✓ | Session identifier for telemetry grouping |
metadata | Model and provider details for telemetry |
Response
{
"status": "blocked",
"threat": {
"type": "pii_exfiltration",
"severity": "critical",
"blocked": true,
"confidence": 0.97,
"violations": [
{
"type": "pii_exposure",
"severity": "medium",
"patterns": ["email_address", "phone_number"],
"confidence": 0.97
}
]
},
"forensics": {
"evidence_hash": "bdbde1558fb883573b0f6e8630ccba1edaf6bed8cb035fa6a48912809149f287",
"has_forensic_payload": true,
"policy_version": "eea811e4f6d3c22074d57b3f9560ad13582e15844a01cd0865481ef7e0b38c56",
"decision_trace_id": "17e797c0-eb5c-4f97-b4a1-2d3e4f5a6b7c"
},
"response": {
"required": true,
"level": "critical",
"escalate": true,
"notify": "security_team"
}
}
Response fields
| Field | Description |
|---|---|
status | "blocked", "flagged", or "clean" |
threat.type | Threat classification |
threat.severity | "low", "medium", "high", or "critical" |
threat.confidence | 0.0–1.0 detection confidence |
threat.violations | Array of specific violation details |
forensics.evidence_hash | SHA-256 of blocked content |
forensics.policy_version | Signed policy bundle hash |
forensics.decision_trace_id | Full trace linkage UUID |
response.escalate | Whether to escalate to security team |
response.notify | Notification target |