What this category's data looks like
This page addresses the category of regular higher education institution admissions filing. Materials come from the admissions plans, filing rules, and admission rosters issued by provincial admissions and examination institutions, as well as the filing and admission progress records returned by institutions. The admissions plan provides the major groups, subject selection requirements, and planned numbers, and is the source of the base figures for statistical purposes; the filing rules specify the filing ratio, tie-breaking ranking, and adjustment scope, and determine which entries in the records are exceptions; the admission roster is the official list of admitted candidates, with dates and signature fields, and is used to check whether the record counts and statistical counts match. The update rhythm advances by filing batch, and a one-week record is formed at the end of each batch, with fields including institution code, major group code, filing count, admission count, withdrawal count, signature, and receipt. When the materials enter this function, the admissions plan serves as the base reference, the filing rules serve as the basis for judgment, and the admission roster serves as the result source document; together, the three support the verification of the progress table and the statistical scope statement.
Configuration and implementation checks
This step is executed by the tool call node in the workflow (the model decides whether to call, and the node executes), outputting the tool call result (the inputs and returns in the call record). The "progress table and statistical scope statement" is then completed by a subsequent AI node or manual verification. Based on the tool description, parameter descriptions, and whether parameters are required, the model determines whether to call the tool this time; when there are no required parameters, the model may fabricate parameter values on its own, and when there are required parameters but no suitable values in the context, it may not call.
| Object | Check | Verification action |
|---|---|---|
| Admissions plan | Check whether the major group codes, subject selection requirements, and planned numbers correspond to the filing ratio in the filing rules, and whether the field dates and version numbers are consistent | Compare the major group codes and planned numbers in the plan table and the rule text one by one, and record the differing rows |
| Filing rules | Check whether the tie-breaking ranking, adjustment scope, and withdrawal conditions are written into the header of the statistical scope statement, and whether exception entries are marked item by item | Open the rule text, copy the exception conditions into the header of the scope statement, and then apply them item by item using the one-week record |
| Admission roster | Check whether the institution code, major group code, admission count, and withdrawal count are equal to the totals in the progress records, and whether signatures and receipts are complete | Use the one-week record to verify the totals, and for rows with discrepancies check the signature field and receipt date |
| Tool call node connections | Check whether the diamond interface is connected to the target tool, whether the node status is executable, and whether the inputs and returns in the call record are complete | Check the diamond connections and node status on the canvas, and run a test to see whether the return contains values that do not exist in the business materials |
Three common errors
- Phenomenon: The name or quantity of the same institution does not match between the plan table and the roster. Cause: The two materials are not synchronized in version; one still uses the old major group code, while the other has already adjusted according to the new rules.
- Phenomenon: There is a discrepancy between the progress record totals and the admission count in the roster, and the discrepant rows have no signature or receipt. Cause: Withdrawal and adjustment entries were not registered in time, and the signature step was left blank, causing the record counts and statistical counts to fail to match.
- Phenomenon: Values that do not exist in the business materials appear in the returned results of the call record, and after multiple tools are chained, the results interfere with each other. Cause: Parameters are not marked as required, the model fabricates parameter values on its own, or the tool description does not specify the boundaries of the returned fields.
How to verify the setup
Acceptance of this function: Run a test with a question that lacks a required parameter, check the inputs and returns in the call record, and confirm whether it asks a follow-up question or calls according to the description; at the same time, check the values of toolChoice and functionCall in the call record to see whether the model first asks a follow-up question and then supplies the parameter.
Business test: Use the one-week record to verify the totals and see whether the scope and exceptions are clearly written; align the fields, dates, and signatures of the three materials—the admissions plan, filing rules, and admission roster—item by item, and annotate the reasons for discrepant rows.
- Open the tool node description field and check whether this description can be understood when taken out alone, and whether it clearly states the purpose, when it should be used, and the returned results.
- Open each parameter configuration one by one, check the parameter description and whether it is required, and run a test with a question that lacks that field to see whether it first asks a follow-up question instead of directly fabricating.