What data belongs to this category
Data for this category comes from local public data disclosure platforms, integrated big data platforms, departmental business databases, and electronic certificate repositories. The update frequency is once per month. Daily generated documents include public data resource catalogs, data disclosure lists, data item descriptions, anonymization rules, interface documents, and update frequency explanations. Covered fields include data item name, disclosure permission, anonymization requirements, interface address, update cycle, and other related items. All content uses structured or semi-structured formats. There is no large volume of unstructured text.
Constraints on Citation Sources and Traceability from These Characteristics
Public data catalogs’ structured nature requires citation traceability to precisely match data item fields. This avoids redundant information interfering with review judgments. Monthly update frequency requires traceability systems to sync with the latest catalog versions. This prevents citation of expired data. Distributed data from multiple sources requires traceability information to clearly mark the source platform or department. This ensures traceability for file review. Anonymization rules require traceability processes to not disclose unanonymized raw data. Only public fields that meet disclosure conditions may be displayed. These characteristics jointly constrain citation configurations to balance accuracy, timeliness, and compliance.
Configuration Settings
| Configuration Item | Recommended Approach | Rationale |
|---|---|---|
quoteMaxToken | 800–1200 token | Matches the average length of public data catalog documents. Prevents truncation of key traceability information such as data item descriptions and anonymization rules |
responseData | Return only structured fields | Public data catalogs are mostly structured content. Filtering unstructured redundant information improves review efficiency |
detail | Enabled | Full display of traceability details such as data item source and update cycle is required to meet compliance requirements for file review |
| Maximum context | First 3 rounds of conversation context | Adapts to multi-round verification scenarios in declaration material review. Value range is determined based on actual testing |
| Single citation count | Top 5 entries | Matches the reasonable data volume referenced in a single file review. Avoids information overload that impairs judgment |
Three Common Mistakes
- Phenomenon: Responses only show citation identifiers without body content. Cause:
responseDatareturn rules are not configured correctly. Only citation metadata is loaded, and valid content is not extracted. - Phenomenon: Citation view function for login-free share links fails. Cause: Versions 4.9.6 and earlier do not sync permission configurations to the share link. Upgrade to FastGPT v4.17.0 and adjust corresponding settings.
- Phenomenon: Answer paragraphs do not display citation source identifiers. Cause: The citation display switch in the
detailconfiguration is not enabled, or knowledge base chunks are not correctly associated with metadata information for corresponding data items.
How to Confirm Proper Configuration
- Upload one public data catalog document. Submit a query that includes data items. Check if the answer includes a clear citation source identifier at the end.
- Generate a login-free share link. Click the view original citation button in the link. Confirm that you can navigate to the detailed description page for the corresponding data item.
- Submit two consecutive related queries. Check that the system can use data catalog information from the previous conversation as a basis for responses.
- Access the application configuration panel. Verify that the settings for
quoteMaxToken,responseData, anddetailmatch the preset configurations.
The values provided here are common starting points. They should be evaluated against the reader’s own samples.
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.