What This Category’s Data Looks Like
Data sources for this category include resin batches and property sheets, injection molding machine operating parameters, mold temperature controller data, workshop MES systems, and dimensional and appearance inspection results. Updates occur every production batch, each mold or material change, and at the end of each shift. Daily documents include injection molding process cards, raw material drying records, mold history logs, first-piece inspection forms, production patrol forms, and material change and machine cleaning records. Fields covered include resin batch number, mold temperature (degrees Celsius), injection molding speed, dimensional tolerance (millimeters), production cycle (seconds), and more.
Constraints Imposed on Tool Calling and Plugins
Data for this category updates frequently and is triggered by production nodes. This requires tool calling to be bound to specific production batches and mold or material change events, to avoid calling expired or unrelated production data. The presence of multiple types of daily documents requires plugins to support parsing files in different formats such as structured tables and log records. Dispersed data sources require plugins to implement multi-API docking to integrate data from injection molding machines, mold temperature controllers, MES systems, and other platforms. Specific unit requirements for fields mean plugins must include built-in format verification rules to ensure parameters passed to tools conform to standard injection molding industry formats. At the same time, quality audit errors can lead to batch defects or safety issues, so tool calling must be configured with failure retry and exception alert logic.
Configuration Settings
These configurations are based on FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
toolChoice capability flag | Set to true or false to match the selected model/API documentation | Declares model support for tool choice; it is not an auto mode. Configure the actual tools in the workflow or integration |
functionCall capability flag | Set to true or false to match the selected model/API documentation | Declares function-calling support; it is not a tool group or function list. Configure the actual callable tools in the workflow or integration |
| Tool calling timeout | 300–600 seconds | Injection molding data collection and multi-document parsing require processing time, to avoid interrupting valid calls due to timeout |
| Data source binding rule | Associate corresponding resin, mold, and injection molding machine data by production batch | Data for this category is updated by batch, ensuring called data fully matches the current audit batch |
| Failure retry count | 2 times | Addresses temporary network fluctuations during injection molding production data collection, avoiding misjudgment of tool calling failures |
| Log output level | Detailed | Fully record parameters, return results, and exception information for tool calls, to facilitate troubleshooting of injection molding quality issues |
Three Common Misconfigurations
- Phenomenon: The tool calling node returns the
Invalid JSON: Bad control charactererror. Cause: When splicing parameters passed to the tool, control characters such as line breaks and tabs in injection molding process cards and drying records are not escaped, resulting in invalid JSON format. - Phenomenon: The tool calling node cannot create connection lines, and no small circle controls for connection are displayed on the interface. Cause: The basic configuration of the tool calling node is not completed, or the edit activation state of the workflow is not enabled, causing interactive controls to be hidden.
- Phenomenon: The MCP-assisted model takes too long to process injection molding audit tasks, or detailed call logs for the MCP service cannot be viewed. Cause: The maximum thinking token count for the model is not limited, only the injection molding-related tool group is not called, or the log output level is not configured as detailed, causing the model to retrieve irrelevant content or lack log information.
How to Confirm Configuration Is Complete
- Perform a tool calling test, check whether the returned JSON format is valid, with no parsing errors caused by control characters.
- Manually trigger tool calling, verify that it can correctly associate resin, mold, and injection molding machine data for the current production batch, and that returned fields include standard injection molding industry parameters.
- Simulate a network fluctuation, check whether the tool calling automatically retries according to the configured retry count, with no abnormal errors.
- View the workflow log panel, confirm that detailed parameters and return results of tool calls have been fully recorded.
The values provided here are common starting points, and should be evaluated against the reader's own samples.