Incidents API

Incidents are automatically created when alert rules fire. They track the lifecycle of an alert event from detection through resolution. Incidents are managed through the dashboard API (not the client API).

INFO

These endpoints are dashboard APIs, authenticated via JWT session (not API tokens). They are used by the Cognipeer Console UI.

List Incidents

GET /api/alerts/incidents

Query Parameters

Parameter	Type	Required	Description
ruleId	string	No	Filter by alert rule ID
status	string	No	Filter by status: open, acknowledged, investigating, resolved, closed
severity	string	No	Filter by severity: critical, warning, info
limit	number	No	Max results (default: 50)
skip	number	No	Pagination offset (default: 0)

Response

{
  "incidents": [
    {
      "_id": "inc_abc123",
      "ruleId": "rule_xyz",
      "ruleName": "High Error Rate",
      "metric": "model_usage_error_rate",
      "threshold": 5,
      "actualValue": 12.3,
      "severity": "critical",
      "status": "open",
      "notes": [],
      "statusHistory": [
        { "status": "open", "changedBy": "system", "changedAt": "2026-03-01T10:00:00.000Z" }
      ],
      "createdAt": "2026-03-01T10:00:00.000Z",
      "updatedAt": "2026-03-01T10:00:00.000Z"
    }
  ],
  "openCount": 3
}

Get Incident

GET /api/alerts/incidents/:incidentId

Response

{
  "incident": {
    "_id": "inc_abc123",
    "ruleId": "rule_xyz",
    "ruleName": "High Error Rate",
    "metric": "model_usage_error_rate",
    "threshold": 5,
    "actualValue": 12.3,
    "severity": "critical",
    "status": "acknowledged",
    "notes": [
      {
        "userId": "user_1",
        "userEmail": "admin@example.com",
        "content": "Investigating upstream provider issues",
        "createdAt": "2026-03-01T10:30:00.000Z"
      }
    ],
    "statusHistory": [
      { "status": "open", "changedBy": "system", "changedAt": "2026-03-01T10:00:00.000Z" },
      { "status": "acknowledged", "changedBy": "user_1", "changedAt": "2026-03-01T10:15:00.000Z" }
    ],
    "createdAt": "2026-03-01T10:00:00.000Z",
    "updatedAt": "2026-03-01T10:15:00.000Z"
  }
}

Update Incident Status

PATCH /api/alerts/incidents/:incidentId

Request

{
  "status": "acknowledged"
}

Valid Statuses

Status	Description
open	Initial state when incident is created
acknowledged	Operator has seen the incident
investigating	Active investigation in progress
resolved	Root cause addressed
closed	Incident fully closed

Status Workflow

open → acknowledged → investigating → resolved → closed
  ↑                                      │
  └──────────── reopen ──────────────────┘

Resolved incidents may be reopened back to open if the issue recurs.

Add Note

POST /api/alerts/incidents/:incidentId/notes

Request

{
  "content": "Identified root cause: upstream provider rate limiting"
}

Parameters

Field	Type	Required	Description
content	string	Yes	Note text (non-empty)

Response

Returns the updated incident with the new note appended.

Severity Levels

Severity is automatically calculated when an incident is created based on the metric type and how far the actual value exceeds the threshold:

Severity	Description
critical	Actual value significantly exceeds threshold
warning	Threshold breached but within moderate range
info	Minor threshold violation

Errors

Status	Description
400	Missing or invalid status value
401	Not authenticated
404	Incident not found
500	Internal server error