Skip to main content

🔄 API Execution Flow

This guide outlines the standard lifecycle for building applications using the DataGOL API. By following this sequence, you can programmatically set up environments, structure data, and leverage AI enrichment without manual intervention.

🗺️ Visual Flowchart​

The following diagram illustrates the dependency and execution order of the DataGOL API suite.

graph TD
    A[🔐 Login] -->|Extract JWT| B[🔑 Create Service Account]
    B -->|Use x-auth-token| C[🏢 Create Workspace]
    C -->|Extract workspaceId| D[📊 Create Table]
    D -->|Extract tableId| E[📝 Add Rows & Columns]
    E --> F{🤖 AI Enrichment?}
    F -->|Yes| G[✨ Create AI Column]
    F -->|No| H[🚀 Ready to Fetch Data]
    G --> H
    H --> I[📥 Get Table Data]

🏗️ Real-Time Example: Automated Lead Management​

Imagine you are building a CRM App that automatically fetches leads and uses AI to categorize them based on their industry.

Step 1: Initialize the Environment​

First, ensure you have a workspace where your table will live.

Endpoint: POST https://be.datagol.ai/noCo/api/v2/workspaces
Goal: Create a space called "Sales CRM".

Step 2: Create the Leads Table​

Structure your data to hold lead information.

Endpoint: POST https://be.datagol.ai/noCo/api/v2/workspaces/:workspaceId/tables
Request Body:

{
  "name": "Leads",
  "isIncludeColumns": true
}

Step 3: Add Data and AI Enrichment​

This is where the magic happens. You add columns for standard data and an AI column to summarize the lead's potential.

Add Standard Column:

{
  "name": "company_description",
  "uiDataType": "LONG_TEXT",
  "uiMetadata": { "title": "Company Description" }
}

Add AI Column (The "Super" Part): This API call enables AI to read the company_description and automatically output a "Lead Score".

{
  "name": "lead_score",
  "uiDataType": "SINGLE_LINE_TEXT",
  "uiMetadata": { "title": "AI Lead Score" },
  "settings": {
    "aiSettings": {
      "enabled": true,
      "prompt": "Based on the company_description, rate this lead from 1-10",
      "model": "gpt-4o-mini"
    }
  }
}

Step 4: Finalize and Fetch​

Once the structure is set, you can push row data. The AI column will populate automatically.

Access Data: Endpoint: POST https://be.datagol.ai/noCo/api/v2/workspaces/:workspaceId/tables/:tableId/data/external
Result: You get a clean JSON response with both your manual inputs and AI-generated insights.

🤖 Machine Readability (AI-First)​

If you are using Claude Code, Cursor, or any other LLM-based developer tool:

  1. Feed the Spec:
  2. Context Injection: Tell the AI: "I want to create a Workspace, then a Table, then an AI column. Follow the flow documented in DataGOL API Flow."
  3. No Manual Work: The AI will use the spec to generate the exact cURL or Fetch commands needed for your specific language.
Pro Tip

You can enable Web Access in the aiSettings if you want the AI column to research company info from the live web!

Was this helpful?