What this category of data includes
Data sources include the national land spatial planning one-map, project approval and supervision platform, land use status database, and ecological protection red line data. Update frequency is irregular. Processed documents include land pre-examination and site selection opinions, construction land planning permits, construction project planning permits, red line drawings, and planning conditions. Data fields cover project site selection scope, land scale, planning control requirements, pre-approval progress, and more. Units include spatial parameters such as hectares and square meters. This data is maintained jointly by the purpose control and planning permission office of the natural resources authority, administrative examination and approval bureaus, and project units.
Constraints for Tool Calling and Plugin Workflows
Multiple data sources with irregular update cycles require tool calling to support connection to multi-source heterogeneous government databases. Incremental synchronization trigger logic must be configured to adapt to non-fixed update rhythms. Diverse document types with inconsistent formats require plugins to parse approval documents and spatial drawings of different formats. Plugins must extract standardized fields from these materials. Data fields include spatial coordinate parameters. Tool calling must support geographic information format processing to avoid field parsing deviations. Pre-approval relies on data from multiple departments. Tool calling must support cross-platform interface linkage to obtain pre-condition information from development and reform, housing and urban-rural development, and other departments.
Configuration Settings
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
toolChoice | auto | Adapt to automatic scheduling scenarios for multi-tool linkage |
functionCall | Specify land pre-examination tool ID | Accurately call the corresponding government data query tool |
| Tool call trigger timing | Automatically trigger after file submission | Match fixed approval process nodes for land pre-examination and planning permits |
| Knowledge base recall limit | Top 8 entries | Adapt to batch recall requirements for multi-source planning data |
| Tool call timeout threshold | 300 seconds | Correspond to average response time for cross-department government data retrieval |
| Single tool call parameter count | No more than 5 | Avoid concurrent conflicts in cross-department interface calls |
Three Common Misconfigurations
- Phenomenon: After enabling the tool call node, knowledge base retrieval returns no results or no planning-related documents. Cause: Context priority configuration conflict exists between tool call and knowledge base retrieval. Tool call occupies the full context window by default, leaving no input space for knowledge base retrieval.
- Phenomenon: When debugging the tool call node, the log shows two complete thought process outputs. Cause: Both automatic thinking and tool call trigger logic are configured in the workflow, leading to repeated triggering of the thinking step.
- Phenomenon: When calling the workflow via API, configured MCP tool execution results cannot be obtained. Cause: The
toolChoiceparameter is not correctly specified as auto or the corresponding tool ID in the API request, so the tool is not scheduled.
How to Confirm Proper Configuration
- Submit a simulated declaration file, check if the tool call node triggers automatically at the preset timing.
- View the workflow execution log in FastGPT v4.17.0, confirm that retrieved documents include exclusive document types related to land pre-examination.
- Call the API to trigger the workflow, verify that returned results include configured MCP tool execution data.
- Test abnormal scenarios, confirm that the workflow returns preset corresponding prompts when tool call errors occur.
The values listed above are common starting points, and should be evaluated against the user'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.