What this category of data looks like
Reference only; verify against the target FastGPT release and deployment documentation.
What constraints do these characteristics impose on forms and interactions
The multi-scenario split document structure requires the userSelect component to display options grouped by service scenario, to avoid recall bias caused by users entering vague questions. Daily updated temporary change data requires configuration that supports real-time knowledge base synchronization, to prevent response content from lagging behind scenarios like flight adjustments and temporary control. The effective time period field requires form interaction to add time range verification, only displaying currently effective response text. The applicable terminal field requires the form to filter knowledge base recall results based on the terminal information entered by the user, reducing irrelevant content returned. For real-time flight-related user inquiries, form interaction must also support linking to the airport's official real-time data interface to supplement dynamic information not covered by the static knowledge base.
Configuration Settings
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
userSelect | Grouped by service scenario, options include flight services, baggage services, facility services, emergency services | Matches the document's scenario classification structure, reduces user input ambiguity |
formInput | Input length limited to 20–200 characters | Covers the length of most passenger inquiry questions, avoids reduced recall efficiency from overly long inputs |
| Single search limit | Top 3 results | Answers for civil aviation airport inquiry scenarios usually focus on three core directions; too many results increase user reading burden |
| Reordering | Sorted by effective time in descending order | Temporary change inquiry content has high update frequency, prioritize displaying the latest effective responses |
| Maximum context | 600 seconds | Passenger inquiries usually revolve around questions related to a single trip; limiting context length to 10 minutes covers the complete inquiry flow |
| Question optimization | Enabled | Civil aviation airport inquiry questions often contain colloquial expressions; optimization improves keyword matching accuracy |
Three Common Mistakes
- Symptom: Unable to select the specified airport knowledge base when calling a tool. Cause: The
userSelectcomponent is not bound to the corresponding scenario's knowledge base group, resulting in no filtering of available knowledge bases during tool calls. - Symptom: Significant interface lag occurs when dragging the
formInputoruserSelectcomponent. Cause: The total number of bound knowledge base documents exceeds 1000, and component lazy loading configuration is not enabled. - Symptom: Responses returned by API calls differ significantly from those in online chat. Cause: Online chat enables question optimization by default, while API calls do not have this option configured, leading to insufficient keyword matching accuracy for colloquial questions.
How to Confirm Proper Configuration
- Navigate to the form editing interface, confirm that the
userSelectcomponent displays options grouped by scenarios such as flight services and baggage services. - Input inquiry content exceeding 200 characters, check if the interface displays a length limit exceeded prompt.
- Initiate a simulated inquiry, verify that recall results are sorted by effective time in descending order.
- Initiate the same inquiry via online chat and API calls respectively, confirm that response contents are basically consistent.
Note: The configuration values listed above are common starting points. They should be evaluated against applicable samples and use cases.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.