API reference

Send Authorization: Bearer <project_token> on every request. Tokens are managed under Settings → API keys. Invalid or missing tokens return 401 or 403.

Ingest a new log entry for a project.

POST /api/projects/{projectId}/logs
Authorization: Bearer {token}
Content-Type: application/json

{
  "message": "User signed in",
  "type": "INFO",
  "criticality": "LOW",
  "source": "auth-service",
  "metadata": {
    "userId": "user_abc",
    "plan": "pro"
  }
}

Body fields

All ingest fields are sanitized server-side: HTML tags and dangerous URL schemes are stripped, control/bidi characters removed, and metadata is limited to a plain JSON object (no __proto__or nested functions). Messages over 8 KB or metadata over 16 KB are rejected.

Success: 201 Created with the log JSON body.

Errors: 400 validation, 401 missing token, 403 unauthorized project/token, 500 server failure.

{
  "id": "...",
  "projectId": "...",
  "message": "User signed in",
  "type": "INFO",
  "criticality": "LOW",
  "status": "ACTIVE",
  "source": "auth-service",
  "metadata": { "userId": "user_abc", "plan": "pro" },
  "timestamp": "2025-06-03T12:00:00.000Z",
  "createdAt": "...",
  "updatedAt": "..."
}

Send up to 100 logs in one request. Same auth and field rules as single ingest. Body may be a JSON array or { "logs": [ ... ] }. Each item may use level (debug, info, warn, error) instead of type.

POST /api/projects/{projectId}/logs/batch
Authorization: Bearer {token}
Content-Type: application/json

{
  "logs": [
    { "message": "Worker started", "level": "info", "source": "worker" },
    { "message": "Disk 90% full", "level": "warn", "metadata": { "pct": 90 } }
  ]
}

Success: 201 Created with accepted count and created log objects.

Create a webhook in Settings → Integration. Each hook gets a unique token in the URL. No Bearer auth — the token in the path identifies the hook.

POST /api/projects/{projectId}/hooks/{hookToken}
Content-Type: application/json

{
  "message": "Zapier trigger fired",
  "level": "info",
  "metadata": { "zap": "new-error" }
}

Arbitrary JSON is accepted; common fields (message, level, source, metadata) are mapped automatically. Success: 201 Created with the log body.

Ingest an OpenTelemetry logs export JSON payload. Severity is mapped to LoggerMan type and criticality.

POST /api/projects/{projectId}/otlp
Authorization: Bearer {token}
Content-Type: application/json

{ "resourceLogs": [ ... ] }

Use the Integration tab for a full example export. Same rate limits as single log ingest apply.

Successful ingest responses include optional usage headers for the project owner's current billing period. When you have used at least 80% of included volume, X-LoggerMan-Usage-Warning is set to approaching_limit; beyond included volume it becomes overage. See Pricing & usage for tier allowances and overage rates.

When per-minute ingest limits are exceeded, LoggerMan returns 202 Accepted with code: "QUEUED" instead of dropping payloads. Higher plans dequeue first: Scale, then Team, then Starter. Queued logs are delivered when capacity frees up — retry with backoff if you see sustained 429 responses.

Ingest endpoints return JSON errors with a machine-readable code field when validation or limits fail:

Retrieve logs for a project (paginated).

GET /api/projects/{projectId}/logs
Authorization: Bearer {token}
page: 0

Headers

Success: 200 OK with a JSON array of log objects.

Unauthorized: 403 with empty array when token does not match the project.

• Keep messages concise; put bulk data in metadata.
• Prefer the SDK — it sets JSON headers and surfaces HTTP errors.
• Do not log secrets, passwords, or raw payment card data.

Machine-readable API description for codegen and contract tests:

• /openapi.json — static spec file
• /api/openapi — same spec via API route

Use it with OpenAPI Generator, Speakeasy, or your own client — point your tool at the URL above and set the server to your LoggerMan origin (Cloud: https://loggerman.dev).