This guide is developed for FastGPT v4.17.0, and covers sharing and embedding configuration for party building learning Q&A scenarios in the social security field.
What this category of data looks like
Data for this category comes from social insurance handling systems, personal benefit records, labor relations and work injury certification systems, and electronic social security card platforms, and is updated daily. Daily associated documents include benefit application forms, payment year proof documents, retirement approval forms, work injury certification materials, and unemployment benefit application materials. Data fields cover benefit type, verification basis, application eligibility, disbursement standards, and more. Units include monetary amount, payment years, and disbursement months. Each individual data entry links to full process information for personal social insurance participation and benefit verification.
What constraints do these characteristics impose on the sharing and embedding process
The data characteristics of this category impose multiple constraints on the sharing and embedding process. The daily updated data source requires that the embedded Q&A system regularly synchronizes the knowledge base to avoid using expired benefit verification information. The complex structure of multiple associated documents requires clear definition of the data source range when sharing, to prevent cross-category data confusion that interferes with party building learning Q&A. Fields involving sensitive information such as amounts and eligibility conditions require permission control configuration after embedding, to avoid unauthorized access. Q&A in the party building learning scenario must strictly match official verification rules. The shared API must support parameterized filtering of compliant data, while also supporting embedding forms for multiple scenarios such as window terminals and mobile terminals.
How to configure settings
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
| Shared link validity period | 7 days | Adapts to temporary sharing needs of party building learning scenarios, avoids permission risks from long-term links |
| Embedded iframe size | 100% width, 600–800 pixel height | Adapts to common page layouts of government service windows and street service halls |
| Knowledge base recall upper limit | Top 6 entries | Matches the high relevance feature of this category of data, avoids excessive redundant information interfering with Q&A |
shareId | Generated per application instance | Used to uniquely identify shared Q&A instances, ensuring independent permission control for embedding scenarios |
| External image zoom toggle | Enabled | Solves the problem of unclear image viewing within embedded iframes, meets actual usage needs |
| API timeout threshold | 30 seconds | Adapts to the response rhythm of social security data queries, avoids long wait times affecting user experience |
Three common misconfigurations
- Phenomenon: After sharing as an API interface, some questions fail to recall knowledge base content, while the debug page returns results normally. Cause: The API call does not correctly carry the
shareIdparameter associated with the knowledge base, or does not specify the permission scope of the corresponding data source. - Phenomenon: Images in the chatbot reply of the embedded webpage can only be viewed within the iframe, and cannot be zoomed externally. Cause: The external jump and zoom option for images in the embedding configuration is not enabled, and the default rendering is limited to the iframe.
- Phenomenon: The locally deployed MCP service encapsulated via the SSE protocol cannot be recognized and called by FastGPT. Cause: The current version of FastGPT only supports stdio process-based MCP service access, and does not adapt to the communication method of the SSE protocol.
How to confirm the configuration is complete
- Call the API interface generated by sharing, enter questions related to social security benefit verification, and check that the returned results are consistent with those on the debug page.
- Send a test question containing images in the embedded chatbot, confirm that clicking the image opens in a new browser tab for zooming.
- Check the size of the chatbot on the embedded page, confirm that it adapts to the current page layout with no overflow or abnormal white space.
- Verify the response time of the API interface, confirm that it meets the waiting requirements of the business scenario, with no obvious timeouts.
The values provided here are common starting points, and 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.