Build a CRUD API Without Code

Build a complete REST API with database operations using Tensorify's visual editor. No Express, no Django, no boilerplate — just drag nodes, connect them, and deploy.

Uses: API Trigger, Switch, DB Find, DB Insert, DB Update, DB Delete, Return Time: ~15 minutes Template: CRUD API

A REST API with four endpoints on a single path:

| Method | Action | Description | |---|---|---| | GET /items | Read | Fetch all items from the database | | POST /items | Create | Insert a new item | | PUT /items?id=123 | Update | Update an existing item by ID | | DELETE /items?id=123 | Delete | Remove an item by ID |

• A Tensorify workspace
• A PostgreSQL database (or any supported database)
• DATABASE_URL environment variable configured

The fastest way is to use the CRUD API template:

1. Go to Templates → APIs → CRUD API
2. Click Use template

Or build from scratch following the steps below.

1. Drag an API Trigger onto the canvas
2. Configure:
   • Protocol: rest
   • Path: /items
   • Allowed Methods: GET, POST, PUT, DELETE
   • Auth Type: bearer-token (recommended for production)
   • Response Mode: use-workflow-response

1. Add a Switch node
2. Set Condition to: {{ api_request.method }}
3. Add cases: GET, POST, PUT, DELETE
4. Set Match Mode to exact

Connect the API Trigger outputs to the Switch input.

1. Add a DB Find node
2. Configure:
   • Table: items
   • Columns: *
   • Limit: 50
3. Connect Switch GET output → DB Find input

1. Add a DB Insert node
2. Configure:
   • Table: items
   • Data: {{ api_request.body }}
3. Connect Switch POST output → DB Insert input

1. Add a DB Update node
2. Configure:
   • Table: items
   • Data: {{ api_request.body }}
   • Where: id = '{{ api_request.query.id }}'
3. Connect Switch PUT output → DB Update input

1. Add a DB Delete node
2. Configure:
   • Table: items
   • Where: id = '{{ api_request.query.id }}'
3. Connect Switch DELETE output → DB Delete input

Add a Return node after each database operation. Connect each DB node's result output to a Return node's input.

For the POST return, you can set the status code to 201 (Created).

Go to Settings → Environment Variables and add:

• DATABASE_URL — your PostgreSQL connection string (e.g., postgresql://user:pass@host:5432/db)

1. Click Deploy and choose cloud or CLI execution
2. Open the Playground to test each endpoint:

# Create an item
curl -X POST https://triggers.tensorify.io/h/YOUR_PATH/items \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "Widget", "price": 9.99}'

# List items
curl https://triggers.tensorify.io/h/YOUR_PATH/items \
  -H "Authorization: Bearer YOUR_TOKEN"

# Update an item
curl -X PUT "https://triggers.tensorify.io/h/YOUR_PATH/items?id=1" \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "Updated Widget", "price": 14.99}'

# Delete an item
curl -X DELETE "https://triggers.tensorify.io/h/YOUR_PATH/items?id=1" \
  -H "Authorization: Bearer YOUR_TOKEN"

To validate incoming data before database operations, add a Code node between the Switch and the DB Insert/Update:

import json

body = json.loads(data)

if not body.get("name"):
    raise ValueError("name is required")
if not isinstance(body.get("price"), (int, float)):
    raise ValueError("price must be a number")

result = body

The AI Agent's error branch catches exceptions and can return appropriate error responses.

• Build an API — Simpler API guide for beginners
• Expressions and Conditions — Master {{ }} bindings
• Webhook Automation — Handle incoming webhooks
• Deploy as an OpenAI Endpoint — Add AI capabilities to your API