Skip to main content

​
Overview

Invenicum integrates with Google’s Gemini AI to provide advanced capabilities like automatic product data extraction from URLs and intelligent inventory assistance through the Venibot chatbot.
Gemini integration is optional but highly recommended for enhanced productivity and automation features.

​
Prerequisites

  • Active Google account
  • Access to Google AI Studio
  • Invenicum account with active session

​
Step-by-Step Setup

1

Get Your Gemini API Key

  1. Visit Google AI Studio
  2. Sign in with your Google account
  3. Navigate to API Keys section
  4. Click Create API Key or Get API Key
  5. Copy the generated API key (format: AIza...)
Keep your API key secure. Never share it publicly or commit it to version control.
2

Configure Integration in Invenicum

  1. Open Invenicum and navigate to Settings → Integrations
  2. Find the Google Gemini AI integration card
  3. Click Configure or tap on the card
  4. In the configuration sheet, enter your API key
​
apiKey
string
required
Your Google Gemini API key from AI Studio. This field is masked for security.
​
model
string
default:"gemini-3-flash-preview"
The Gemini model to use. Recommended: gemini-3-flash-preview for optimal performance.
3

Save and Verify

  1. Click Save to store your configuration
  2. The integration status should change to Connected ✓
  3. You’ll see a confirmation message
The API key is stored securely on the backend and never exposed in the client.
4

Enable AI Features

  1. Navigate to Settings → General Settings
  2. Find the AI Features toggle
  3. Enable it to activate Gemini-powered features
The AI toggle will only be available after successfully connecting Gemini integration.

​
Available AI Features

Once configured, you can use these Gemini-powered features:

​
Magic Assistant (URL Extraction)

Automatically extract product information from URLs:
  1. Create or edit an asset
  2. Click AI Magic Assistant
  3. Paste a product URL (Amazon, eBay, etc.)
  4. Click Start Magic
  5. AI will fill fields like name, description, price, and more
The AI extraction endpoint is /ai/extract and can extract custom fields defined in your asset types.

​
Venibot Chatbot

Ask questions about your inventory:
  • “What’s in stock?”
  • “Show me active loans”
  • “Generate inventory report”
Access Venibot from the dashboard or inventory screens.

​
Configuration Files

The Gemini integration is defined in: Integration Definition: lib/core/utils/constants.dart:25-42 
IntegrationConfig(
  id: 'gemini',
  name: 'Google Gemini AI',
  description: 'Connect with Google\'s Gemini AI',
  fields: [
    ConfigField(
      id: 'apiKey',
      type: ConfigFieldType.text,
      label: 'API Key',
      helperText: 'Get it at https://aistudio.google.com/',
    ),
    ConfigField(
      label: 'Model',
      type: ConfigFieldType.text,
      helperText: 'Default: gemini-3-flash-preview',
    ),
  ],
)
AI Service: lib/data/services/ai_service.dart:1-53

​
Troubleshooting

  • Verify your API key is correct
  • Check that the key hasn’t expired in Google AI Studio
  • Ensure you saved the configuration
  • Try disconnecting and reconnecting the integration
  • Verify Gemini integration is connected
  • Check your internet connection
  • Ensure the URL is publicly accessible
  • Some websites may block automated extraction
  • Review error message in the app for specific details
The AI features toggle requires:
  • Gemini integration to be connected first
  • Navigate to Settings → Integrations → Google Gemini
  • Complete the setup steps above
Google AI Studio has usage limits:
  • Check your quota in AI Studio dashboard
  • Consider upgrading your Google AI plan
  • Rate limits reset daily

​
Security Best Practices

Never share your API key. It provides direct access to your Google AI resources and usage quota.
  • Store API keys securely using the integration settings (not in code)
  • Rotate keys periodically in Google AI Studio
  • Monitor usage in AI Studio to detect unauthorized access
  • Use environment-specific keys for development vs production

Build docs developers (and LLMs) love

Get started for freeTalk to us