What the data for this category looks like
Data sources for this category include the National Land Spatial Basic Information Platform, natural resource survey and monitoring data, population and economic statistical data, and remote sensing images. Daily produced documents cover overall land spatial plans, detailed plans, special plans, planning implementation evaluation reports, and public opinion summaries. Data is updated annually, and includes fields such as decimal longitude and latitude coordinates, construction land area (hectares), population scale (ten thousand people). The data has three types of formats: spatial vector, attribute table, and text reviews.
What constraints these characteristics impose on the tool calling and plugins workflow
The multi-source and heterogeneous data characteristics of this category require tool calling to cover three types of plugins: spatial analysis, statistical accounting, and text parsing. The annually updated data source feature requires verifying the data version before each call to avoid using expired planning baselines. Regional-level planning scale attributes require tool calling to support batch spatial data processing, with the input and output volume of a single call matching the field scale of planning documents. The large volume of public opinion summary text requires configuring a tool calling link adapted to long texts, while ensuring that tool calling results can be passed to subsequent conversation modules for planning scheme adjustment and inquiry response.
Configuration Settings
Reference only; verify against the target FastGPT release and deployment documentation.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
toolChoice | auto | Adapt to automatic tool selection requirements across different planning scenarios |
functionCall | Preset tool groups according to planning scenarios | Cover full-process tool calling for spatial analysis, statistical accounting, and text parsing |
| Single tool call timeout | 600 seconds | Spatial analysis tools require long processing cycles for remote sensing images or multi-source data |
| Maximum number of return entries per tool call | Top 10 | Adapt to batch processing and display requirements for planning data |
| Tool call result output format | JSON format, including spatial coordinates and planning indicator fields | Adapt to structured storage of planning data and subsequent module calling requirements |
| Global variable update trigger condition | After tool call completion and verification passes | Prevent ineffective variable values from affecting subsequent planning calculation logic |
Three Common Misconfigurations
- When passing the
chatIdparameter during API calls, the corresponding session identifier is not displayed in the conversation log, and the tool call result is not bound to the specified session. This occurs because the session parameter binding configuration is not enabled in the FastGPT backend, or the request format does not meet platform requirements. - Global variables are not updated after tool call completion, and subsequent conversation modules still use old variable values. This occurs because the variable update logic is not mounted to the tool call success callback node, or the variable scope is set to valid only for a single call.
- An unauthorized API key error with
code:514is returned during tool calls. This occurs because the configured API key is not bound in the FastGPT plugin management, or the key permissions do not cover the currently called tool interface.
How to Verify Proper Configuration
- Initiate a simulated planning data tool call, check whether the returned result format matches the preset JSON structure, including spatial coordinates and planning indicator fields.
- Pass a test
chatIdparameter to initiate an API call, check whether the corresponding session identifier is displayed in the conversation log to confirm correct parameter receipt. - Trigger a tool call and update global variables, initiate a subsequent conversation, check whether variables use updated values.
- Configure a test API key, initiate a tool call, confirm that no
unAuthApiKeyrelated errors are returned.
The values provided here are common starting points and should be evaluated against relevant local 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.