Reference only; verify against the target FastGPT release and deployment documentation.
What the data for this category looks like
Coal transportation and sales data mainly comes from transportation dispatch systems, weighing ticket systems, coal quality testing laboratory systems, contract management platforms, and transportation scheduling systems. Update rhythms are divided into three categories: real-time (weighing ticket data is generated during weighing), daily (updates coal yard inventory and daily shipping plans), and monthly (generates settlement reconciliation reports). The document structure of a single data record includes fields such as contract number, shipping date, coal type, net calorific value (as received basis), actual shipping weight, carrier entity, arrival station, and settlement unit price. Units include tons, kcal/kg, yuan/ton, and other standard units.
Constraints Imposed on Tool Calling and Plugins
Multiple data sources with varying update rhythms require tool calls to adapt to different interface timeout thresholds. Real-time data interfaces need short timeouts, while batch report interfaces can use longer timeouts. Core fields have specific units and association rules, requiring plugins to include built-in field verification and mapping logic to avoid unit mismatches or data association errors. The number of fields per single data record is moderate but closely cross-data-source associated, requiring tool calls to support parameter passing and cross-plugin linkage. Without this support, data fragmentation will occur. Different scenarios have different data update frequencies, requiring plugins to support on-demand call triggering to adapt to the different needs of real-time queries and batch statistics.
How to Set Configurations
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
toolChoice | auto, change to none when only a single tool is specified | Adapt to daily office scenarios with multi-plugin linkage for coal transportation and sales, automatically match corresponding business plugins |
functionCall | Enabled and bound to no more than 3 transportation and sales-specific plugins | Avoid excessive plugins causing context overflow, align with the field count range of single transportation and sales data records |
| Interface timeout duration | 60-120 seconds | Adapt to response durations of real-time interfaces such as weighing ticket queries and coal quality detection data pulling |
| Parameter mapping rules | Associate weighing tickets, coal quality, and settlement data by contract number | Match core association fields of coal transportation and sales data to ensure correct parameter passing |
| Log recording toggle | Enable full-link workflow logs | Facilitate troubleshooting of log loss issues in cross-plugin calls |
| Batch call limit | First 20 pending transportation orders | Avoid interface timeouts caused by excessive data volume in a single call |
Three Common Configuration Errors
- Empty conversation log fields after calling cross-application plugins: Full-link log recording configuration for the workflow is not enabled, only local logs for a single plugin are activated.
- Failed parameter passing during tool calling: Parameter mapping rules are not configured according to the core association fields of coal transportation and sales, causing fields output by the source plugin to be unrecognizable by the target plugin.
- Code execution plugin runtime error: The data analysis library required for coal transportation and sales scenarios is not installed in the plugin environment, or its version is incompatible with existing data formats.
How to Verify Correct Configuration
- Trigger a tool call for a single transportation order query, check if the fields returned by the plugin include core coal transportation and sales fields such as contract number, net calorific value (as received basis), and actual shipping weight.
- Submit a parameter passing test request to confirm that the target plugin can correctly receive and use the transportation order association data output by the source plugin.
- View the workflow log panel to confirm that each step of cross-plugin calls has complete conversation records and return results.
- Run a code execution plugin test to confirm that the specified data analysis library can be normally loaded and used to process transportation and sales related data.
The values provided are common starting points and should be measured 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.