API & Sending Data

Beyond uploading CSV files, you can send event data to Flow Myna programmatically—straight from your applications, ETL pipelines, and integrations—for continuous, real-time process monitoring.

Why Use the API?

CSV upload is perfect for analyzing historical data. The API is for keeping Flow Myna up to date as your process runs:

• Push events the moment they happen in your systems
• Monitor processes continuously instead of in batches
• Combine live data with historical CSV imports in the same dataset

If you can emit an event when something happens—an order is placed, a ticket is resolved, an invoice is paid—you can stream your process into Flow Myna.

Step 1: Create an API Key

1. Go to Workspace → Data → API Keys (under the Integrations section)
2. Click Create API Key
3. Enter a name (e.g., "Production ETL")
4. Choose a dataset: Create new (default, auto-named) or Use existing
5. Copy the key immediately—it's only shown once

The dataset you choose is where events from this key will land.

Step 2: Authenticate

Include your API key in the Authorization header on every request:

Authorization: Bearer fm_live_xxxxx

Or use the custom header:

X-FlowMyna-Api-Key: fm_live_xxxxx

The base URL is https://api.flowmyna.com/v1.

Step 3: Send Events

The simplest call records a single event and the objects it touches. Objects are auto-created if they don't exist yet—the dataset is determined by your API key.

curl -X POST 'https://api.flowmyna.com/v1/event' \
  -H 'Authorization: Bearer fm_live_your_key_here' \
  -H 'Content-Type: application/json' \
  -d '{
    "event": "Order Placed",
    "timestamp": "2024-01-15T10:30:00Z",
    "objects": [
      {"type": "Order", "id": "ORD-123"},
      {"type": "Customer", "id": "CUST-456"}
    ],
    "properties": {
      "total": 149.99,
      "items_count": 3
    }
  }'

That event is now in Flow Myna and will appear in your process map.

Usage Patterns

Pattern 1: Events Only (Simplest)

Just send events—objects are created automatically, and you can include object properties inline (they're merged if the object already exists):

curl -X POST '.../v1/event' \
  -d '{
    "event": "Ticket Created",
    "objects": [
      {"type": "Ticket", "id": "TKT-123", "properties": {"priority": "high"}}
    ]
  }'

Pattern 2: Upsert Objects, Then Record Events (Richer Data)

For processes with rich object data, pre-register objects with their properties, then reference them in events:

# Register the customer once
curl -X POST '.../v1/object/upsert' \
  -d '{
    "type": "Customer",
    "id": "CUST-456",
    "properties": { "name": "Jane Doe", "tier": "gold", "lifetime_value": 5420.00 }
  }'

# Then record events that reference it
curl -X POST '.../v1/event' \
  -d '{
    "event": "Order Created",
    "objects": [
      {"type": "Order", "id": "ORD-789"},
      {"type": "Customer", "id": "CUST-456"}
    ]
  }'

/object/upsert uses upsert semantics—properties are merged, not replaced—so you can enrich an object over time.

Pattern 3: Batch (ETL & Data Pipelines)

For historical imports or scheduled syncs, send many records in one request (up to 100 per call) via /event/batch and /object/batch:

curl -X POST '.../v1/event/batch' \
  -d '{
    "events": [
      { "event": "Case Opened",  "timestamp": "2024-01-01T09:00:00Z", "objects": [{"type": "Case", "id": "CASE-001"}] },
      { "event": "Case Closed",  "timestamp": "2024-01-03T16:00:00Z", "objects": [{"type": "Case", "id": "CASE-001"}] }
    ]
  }'

Endpoints at a Glance

Client SDKs

Prefer your own language to raw HTTP? Official SDKs handle auth, batching, and retries for you:

• Python
• Node.js / TypeScript
• .NET
• Ruby

Each wraps the same endpoints above. Contact us for installation details and the latest versions.

Next Steps

• What Data Do You Need? - The events, timestamps, and IDs to send
• Uploading Your Data - Import historical data via CSV alongside the API
• Object-Centric Approach - How objects and events relate