### Answer A 400 error typically indicates an issue with the request payload or JSON format. Here are the most common causes: 1. Function parameters format issue If you're using custom functions, the parameters must follow the function calling format. [(1) ]() 2. JSON payload validation problems When updating configurations (like prompt trees) , a 400 Bad Request usually means there's a problem with the JSON payload you're sending. Use a JSON validator to ensure it's properly formatted. [(2) ]() 3. Imported agent formatting If you imported an agent, imported agents sometimes have incorrect JSON format. Building from scratch can help isolate whether this is the issue. 4. General tools configuration If you're configuring tools, ensure they match the required schema. For example, certain tool types require specific properties like

cal_api_key

,

event_type_id

, or

number

depending on the tool type. [(3) ]() Could you share more details about what specifically you're configuring in the AI QA cohort setup? That would help narrow down the exact cause.

​ To learn more, see the sources I considered: [[1] Status code 400](https://discord.com/channels/1186424376300679209/1441197468174782577) [[2] Request failed with status code 400](https://discord.com/channels/1186424376300679209/1379532913703391384) [[3] Discord message from @neoxsaibot on 2024-11-06](https://discord.com/channels/1186424376300679209/1294891368505937952/1303866026223730688) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.