You can now create custom API connectors for any data source directly from the AI Agent chat interface. No separate forms, no coding — just describe what you need, and the AI Agent handles the rest.
How it works
Ask the AI Agent — open the AI Agent chat and say something like "Create a custom connector for [platform name]."
Fill in the details — provide the data source name, authentication credentials, and optionally a link to the vendor's API documentation and any specific requirements.
Let the AI build it — the Agent sets up an isolated sandbox, reads the API docs, generates the integration, tests it, and deploys it to your workspace. You'll see real-time progress in the chat.
Explore your data with Discovery API
Once the connector is live, you can explore available endpoints and run test queries right from the chat:
"Show available data for my [platform] connection via Discovery API"
"What endpoints are available in my custom integration?"
"Run a test query to get data from [specific endpoint]"
The AI Agent automatically generates documentation for all discovered API endpoints.
Key benefits
Chat-first experience — everything happens inside the AI Agent, no context switching.
No technical requirements — create integrations without coding knowledge or API expertise.
Rapid implementation — connect to new data sources in minutes, not days.
Secure sandbox — all implementation runs in isolated environments with credential protection.