Skip to main content

​
Overview

AI-enhanced passes analyze your API operations and optimize the generated server for better AI agent interaction. These passes are optional and consume credits.

​
Available passes

​
Parameter curation

Analyzes each operation’s parameters and identifies fields that AI agents shouldn’t set — such as server-generated timestamps, auto-incremented IDs, and internal tracking fields. These are removed from the tool’s parameter list, giving the AI agent a cleaner interface. Before: 23 parameters including created_at, updated_at, internal_id, request_id After: 15 user-controllable parameters

​
Description optimization

Rewrites parameter and tool descriptions to be LLM-friendly. Clear, concise descriptions help AI agents select the right tool and provide correct parameters.

​
Schema optimization

Makes parameter schemas more token-efficient by simplifying complex nested structures while preserving semantic meaning. Reduces the token cost of tool definitions sent to the AI model.

​
How to use

In the dashboard:
  1. Upload your spec and generate the base server (free)
  2. In the Generation tab, enable the AI passes you want
  3. Configure which operations to optimize (all or selected)
  4. Click Generate — credits are consumed based on operation count

​
Credits

AI passes consume credits based on the number of operations and parameters processed. See Pricing for details. Base generation is always free. AI passes are entirely optional.