Bid Document and Quotation Preparation for SMT Mounting and Electronic Assembly: Tool Calling and Plugins

Data for this category comes from BOMs, Gerber files, mounting coordinate files provided by customers, as well as MES systems, mounting machines

What Data for This Category Looks Like

Data for this category comes from BOMs, Gerber files, mounting coordinate files provided by customers, as well as MES systems, mounting machines, SPI/AOI testing equipment, reflow oven temperature recorders, and solder paste and component batch ledgers. Data is updated in real time synchronously with each work order and each production line changeover. Daily documents include BOMs and coordinate files, changeover sheets, stencil records, furnace temperature curves, first article confirmation reports, SPI/AOI defect reports, maintenance records, and traceability labels. Data fields include part numbers, reference designators, component specifications, mounting coordinates, furnace temperature parameters, and defect point coordinates. Units include millimeters, degrees Celsius, batch numbers, and other standard units.

What Constraints These Characteristics Impose on Tool Calling and Plugins

This category of data has strong dynamics and diverse formats. Tool calling must support real-time pulling of the latest work order, changeover, and equipment data, to avoid static caching that causes outdated quotations and document content. The existence of multiple document types requires plugins to adapt to parsing multiple formats such as BOMs, coordinate files, and testing reports, to accurately extract core fields including part numbers, reference designators, and defect points. Data fields include specific units, so tool calling must strictly match field formats and units to ensure numerical accuracy in bid documents and quotations. Bid quotations must also be associated with cost-related data such as component batches and furnace temperature parameters. Plugins must support cross-system data integration and splicing to meet the multi-dimensional information requirements of bid documents.

Configuration Settings

The following configurations are based on FastGPT v4.17.0.

Configuration ItemRecommended SettingRationale
Workflow/API tool selectionautoAutomatically adapts to multi-type tool calling logic, meeting needs for scenarios such as BOM parsing and data pulling
Workflow/API function-call configurationenabledAllows plugins to receive structured function call requests, accurately transmitting core business data such as part numbers and coordinates
Timeout Duration300–600 secondsProcessing multi-format document parsing and cross-MES system data pulling requires a long duration, to avoid task interruptions
Plugin Input FormatSupports mixed JSON and structured textAdapts to structured data from BOMs and text parsing of coordinate files, flexibly transmitting business parameters
Result Return LimitTop 10 entriesBid documents and quotations only require core business fields; too many entries will increase document redundancy and computational load
Image URL RenderingEnable direct displaySupports embedding defect images from SPI/AOI reports in bid documents, without additional downloading or conversion

Three Common Mistakes

  • Symptom: Unable to select preset variables in the JSON input box, with input content format errors or empty content. Cause: The plugin's variable binding configuration is not enabled, and the JSON input box is not associated with system variables or business variables.
  • Symptom: URL image addresses are only displayed as plain text, and cannot be directly rendered in bid documents. Cause: The plugin's image URL direct rendering configuration is not enabled, or image access permission rules are not correctly configured.
  • Symptom: Tool calling tasks time out, returning a timeout error status code. Cause: A timeout duration adapted to multi-document parsing is not set, and waiting duration is insufficient when pulling data across MES systems.

How to Confirm Configuration Is Correct

  • Manually trigger tool calling, upload test BOM and coordinate files, and verify whether returned fields include core business information such as part numbers and reference designators.
  • Configure the JSON input box, bind preset business variables, and confirm that variables can be correctly substituted into plugin logic after entering test content.
  • Upload test data containing image URLs, and check whether generated bid documents directly display image content.
  • Adjust the timeout duration configuration, run long-duration tasks, and confirm that tasks can be completed within the set duration without errors.

The values provided here are common starting points, and should be evaluated against the reader's own samples.