Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.mnemom.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Safe House API reference

The Safe House API covers six functional areas: configuration, quarantine management, observability and metrics, pattern and intelligence management, canary credentials, and compliance exports. All endpoints require a Bearer token or API key unless otherwise noted. Base URL: https://api.mnemom.ai

​
Configuration

Control how Safe House behaves — globally for the org, per-agent, or in bulk.
MethodEndpointDescription
GET/v1/safe-house/configRetrieve org-level Safe House defaults (thresholds, enforcement mode, enabled threat types)
PUT/v1/safe-house/configUpdate org-level defaults — applies to all agents that don’t have a per-agent override
GET/v1/agents/:id/safe-house/configRetrieve per-agent Safe House config (shows effective config after inheritance from org)
PUT/v1/agents/:id/safe-house/configUpdate per-agent config — overrides org defaults for the specified fields only
POST/v1/safe-house/config/bulk-applyApply a config patch to multiple agents at once
Retrieve org-level config: 
curl https://api.mnemom.ai/v1/safe-house/config \
  -H "Authorization: Bearer $TOKEN"

{
  "enforcement_mode": "quarantine",
  "thresholds": {
    "bec_fraud":          { "warn": 0.45, "block": 0.80 },
    "prompt_injection":   { "warn": 0.40, "block": 0.75 },
    "indirect_injection": { "warn": 0.50, "block": 0.85 },
    "social_engineering": { "warn": 0.50, "block": 0.80 },
    "agent_spoofing":     { "warn": 0.45, "block": 0.75 },
    "hijack_attempt":     { "warn": 0.55, "block": 0.85 },
    "data_exfiltration":  { "warn": 0.45, "block": 0.80 },
    "privilege_escalation":{ "warn": 0.45, "block": 0.80 }
  },
  "session_risk_escalation_threshold": 0.70,
  "canary_auto_create": false
}
Update a single agent’s config: 
curl -X PUT https://api.mnemom.ai/v1/agents/mnm-550e8400-e29b-41d4-a716-446655440000/safe-house/config \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "thresholds": {
      "bec_fraud": { "warn": 0.50, "block": 0.88 }
    }
  }'
Bulk-apply a config patch: 
curl -X POST https://api.mnemom.ai/v1/safe-house/config/bulk-apply \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_ids": ["mnm-550e8400-e29b-41d4-a716-446655440000", "mnm-0b3f2a1c-d4e5-4f60-b7a8-9c0d1e2f3a4b"],
    "patch": {
      "enforcement_mode": "block"
    }
  }'

​
Quarantine management

Quarantined turns are held pending human review. Reviewers can release (with or without a false-positive flag) or confirm as a genuine threat.
MethodEndpointDescription
GET/v1/safe-house/quarantineList quarantined items — filter by status, agent_id, threat_type, date range
GET/v1/safe-house/quarantine/:idRetrieve a single quarantine record with full evaluation detail
DELETE/v1/safe-house/quarantine/:idDelete a quarantine record (admin only; irreversible)
POST/v1/safe-house/quarantine/:id/releaseRelease the quarantined turn to the agent; optionally mark as false positive
POST/v1/safe-house/quarantine/:id/reportConfirm the quarantined turn as a genuine threat
List open quarantine items: 
curl "https://api.mnemom.ai/v1/safe-house/quarantine?status=pending&limit=20" \
  -H "Authorization: Bearer $TOKEN"
Release with false-positive flag: 
curl -X POST https://api.mnemom.ai/v1/safe-house/quarantine/qid_7f3a9b2c/release \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "is_false_positive": true,
    "note": "Verified legitimate wire transfer request from CFO"
  }'
Confirm as genuine threat: 
curl -X POST https://api.mnemom.ai/v1/safe-house/quarantine/qid_7f3a9b2c/report \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "note": "Confirmed BEC attempt — forwarded to security team"
  }'

​
Query & observability

Query the full evaluation history, aggregate metrics, and access a live SSE stream for real-time monitoring.
MethodEndpointDescription
GET/v1/safe-house/evaluationsFull evaluation log — filter by agent_id, verdict, threat_type, from, to, min_risk
GET/v1/safe-house/metrics/summaryAggregated counts: total evaluations, block rate, warn rate, false positive rate
GET/v1/safe-house/metrics/timeseriesTime-bucketed metrics for charts — specify bucket (hour, day, week)
GET/v1/safe-house/metrics/threatsTop threat types by volume and confidence over a time window
GET/v1/safe-house/feedSSE stream of live Safe House events — connect once and receive events as they happen
GET/v1/safe-house/sessionsList active sessions with elevated session risk (medium or high)
Query evaluations with filters: 
curl "https://api.mnemom.ai/v1/safe-house/evaluations?agent_id=mnm-550e8400-e29b-41d4-a716-446655440000&verdict=block&from=2026-03-01T00:00:00Z&limit=50" \
  -H "Authorization: Bearer $TOKEN"
Get summary metrics: 
curl "https://api.mnemom.ai/v1/safe-house/metrics/summary?from=2026-03-01T00:00:00Z&to=2026-03-30T23:59:59Z" \
  -H "Authorization: Bearer $TOKEN"

{
  "period": { "from": "2026-03-01T00:00:00Z", "to": "2026-03-30T23:59:59Z" },
  "total_evaluations": 14832,
  "block_count": 47,
  "block_rate": 0.0032,
  "warn_count": 312,
  "warn_rate": 0.021,
  "quarantine_count": 89,
  "false_positive_count": 12,
  "false_positive_rate": 0.135,
  "top_threat_type": "prompt_injection"
}
Connect to the live SSE feed: 
curl -N https://api.mnemom.ai/v1/safe-house/feed \
  -H "Authorization: Bearer $TOKEN" \
  -H "Accept: text/event-stream"
The feed emits safe_house.evaluation.*, safe_house.canary.*, safe_house.session.*, and safe_house.campaign.* events as they occur. Reconnect with Last-Event-ID to replay missed events (replays up to 10 minutes back).

​
Patterns & intelligence

Manage the threat pattern library and retrieve adaptive threshold recommendations.
MethodEndpointDescription
GET/v1/safe-house/patternsList active and candidate threat patterns — filter by status, threat_type
POST/v1/safe-house/patternsSubmit a candidate pattern for review and potential promotion
GET/v1/safe-house/threshold-suggestionsAdaptive threshold recommendations based on your false-positive and miss rate
List active patterns for a threat type: 
curl "https://api.mnemom.ai/v1/safe-house/patterns?status=active&threat_type=bec_fraud" \
  -H "Authorization: Bearer $TOKEN"
Get threshold suggestions: 
curl https://api.mnemom.ai/v1/safe-house/threshold-suggestions \
  -H "Authorization: Bearer $TOKEN"

{
  "suggestions": [
    {
      "threat_type": "bec_fraud",
      "scope": "agent",
      "agent_id": "mnm-550e8400-e29b-41d4-a716-446655440000",
      "current_warn": 0.45,
      "suggested_warn": 0.55,
      "current_block": 0.80,
      "suggested_block": 0.88,
      "rationale": "False positive rate 18.4% over 30 days — above 15% target. Raise warn threshold to reduce noise.",
      "confidence": "high"
    }
  ]
}
Submit a candidate pattern: 
curl -X POST https://api.mnemom.ai/v1/safe-house/patterns \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "threat_type": "bec_fraud",
    "pattern": "(?i)urgent.*wire.*ceo|executive.*transfer.*now",
    "description": "CEO wire fraud variant with transposed urgency/authority order",
    "source": "user_submission"
  }'
Submitted patterns enter candidate status. The arena evaluation pipeline tests them against labeled benign and malicious message sets. Patterns that exceed precision/recall thresholds are promoted to active.

​
Canary credentials

Canary credentials are honeypot API keys, tokens, or other secrets deliberately planted in the agent’s context. If an attacker extracts and uses them, Safe House detects the use and fires a safe_house.canary.triggered event.
MethodEndpointDescription
POST/v1/safe-house/canariesCreate a canary credential and associate it with an agent
GET/v1/safe-house/canaries?agent_id=List canaries for an agent
GET/v1/safe-house/canaries/:id/statusCheck whether a specific canary has been triggered
Create a canary: 
curl -X POST https://api.mnemom.ai/v1/safe-house/canaries \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "mnm-550e8400-e29b-41d4-a716-446655440000",
    "type": "api_key",
    "label": "Honeypot AWS key — do not use",
    "inject_into": "system_prompt"
  }'

{
  "canary_id": "can_f9e2a01b",
  "credential": "AKIAFAKE00HONEYPOT01",
  "type": "api_key",
  "agent_id": "mnm-550e8400-e29b-41d4-a716-446655440000",
  "status": "active",
  "triggered": false,
  "created_at": "2026-03-30T12:00:00Z"
}
The credential value is returned only at creation time. Safe House monitors for its appearance in outbound requests or inbound message content. Check canary status: 
curl https://api.mnemom.ai/v1/safe-house/canaries/can_f9e2a01b/status \
  -H "Authorization: Bearer $TOKEN"

​
Special endpoints

​
Cross-Agent campaign detection

List detected attack campaigns — groups of related attacks targeting multiple agents from the same infrastructure. 
curl "https://api.mnemom.ai/v1/safe-house/campaigns?status=active" \
  -H "Authorization: Bearer $TOKEN"

{
  "campaigns": [
    {
      "campaign_id": "camp_b3c9d4a1",
      "status": "active",
      "threat_type": "bec_fraud",
      "affected_agents": ["mnm-550e8400-e29b-41d4-a716-446655440000", "mnm-0b3f2a1c-d4e5-4f60-b7a8-9c0d1e2f3a4b"],
      "agent_count": 2,
      "first_seen": "2026-03-30T16:50:00Z",
      "last_seen": "2026-03-30T17:20:00Z",
      "similarity_score": 0.92
    }
  ]
}

​
EU AI Act compliance export

Export Safe House evaluation data in EU AI Act Article 50 compliance format. 
curl "https://api.mnemom.ai/v1/compliance/safe-house-report?from=2026-01-01T00:00:00Z&to=2026-03-31T23:59:59Z" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Accept: application/json"
Returns a structured compliance report covering all evaluation decisions, blocked/quarantined turns, false positive resolutions, and configuration change audit records within the requested window. Supports Accept: text/csv for spreadsheet-compatible export.

​
Error responses

All Safe House endpoints return standard Mnemom error objects: 
{
  "error": "Quarantine item not found",
  "type": "not_found",
  "quarantine_id": "qid_7f3a9b2c"
}
HTTP StatusMeaning
400Invalid request body or parameters
401Missing or invalid authentication
403Insufficient permissions for the requested operation
404Resource not found
429Rate limit exceeded
500Internal server error

​
See also