The AI agent (create/edit) panel features two main sections: the left side for configuring the agent and the right side for execution and performance preview. To see how import requests will look ...
Performing a high-volume real-time webhook flow in debug mode may cause increased connection queue size (253k to 152k) causing the flow to run slowly. This can lead to flow failure and can also aff...
You can enter mock response data when configuring imports instead of executing a call to retrieve live response data from an import step. The mock response data that you provide for an import appl...
Tool output defines the data your tool returns after it runs. It determines what information is passed to the next step or back to the calling system. Configure your output to ensure the returned ...
This article answers common questions about execution logs and helps you understand quickly how to use them to monitor, trace, and troubleshoot your flows. What are execution logs? Execution log...
Input control defines how you pass data from a flow to an AI agent. You use it to build clear, reliable inputs from flow records, plain text, and supported files, while staying within model capabi...
Maintaining a tool after it's been created is very straightforward. The Tool builder tab lists all the tools you've created, including: Name: Displays the tool name you gave when you created i...
You can debug a Tool in test mode to ensure the Tool processes your records as expected before using the Tool. When you run a Tool in test mode, you can review the processing behaviour of each ste...
Output control determines how an AI agent returns data to your flow. Configure the output type to ensure the agent’s response is predictable, validated, and easy to map in downstream steps. You se...
MCP prompts are reusable, server-defined message templates that resolve into a sequence of chat messages. You can use prompts to standardize how the model receives instructions and reuse that patt...