What this category's data looks like
Daily materials of a peer review organization mainly consist of review invitations, expert review comments, and author revision notes. The three often arrive in separate batches, and their update cadence follows manuscript circulation and review rounds. Review invitations record the invitee, review deadline, and receipt; expert review comments record the comment body, review date, and signature; author revision notes record point-by-point responses, revision dates, and version identifiers. When the materials enter the Q&A intake and missing-item follow-up feature, review invitations and expert review comments are used to identify review status and comment completeness, while author revision notes are used to compare whether response items and versions are in sync. Field characteristics center on four categories: name, quantity, date, and signature/receipt. The same record often shows mismatched name, quantity, or date across two materials, unsynchronized versions, or missing signatures and receipts. These materials serve as context input for the tool calling node; the node decides whether to call based on the tool description and parameter descriptions, and the returned result is used by subsequent nodes or manual verification for completion.
Configuration and implementation checks
This step is executed by the tool calling node in the workflow (the model decides whether to call, and the node executes), outputting the tool call result (the input parameters and return in the call record). "Intake records and completion questions" are then completed by subsequent AI nodes or manual verification. The model decides whether to call the tool this time based on the tool description, parameter descriptions, and whether parameters are required; when there are no required parameters, the model may fabricate parameter values on its own, and when there are required parameters but no suitable value in the context, it may not call.
| Object | Check | Verification action |
|---|---|---|
| Review invitation | Fields include invitee, review deadline, and receipt status, and must match the date and signature in the expert review comments | Open the tool node description fields and confirm they state the purpose, when to use it, and the return result; test-run with a question missing the receipt field to see whether it asks a follow-up first instead of fabricating directly |
| Expert review comments | Fields include comment body, review date, and signature, and must be consistent with the version identifier in the author revision notes | Open each parameter configuration one by one, check the two items of parameter description and whether it is required, and confirm the required items correspond to fields that actually exist in the business materials |
| Author revision notes | Fields include point-by-point responses, revision dates, and version identifiers, and must be consistent with the name and quantity in the review invitation and expert review comments | Check the diamond connectors and node status on the canvas to confirm they are connected to the target tool; check whether the return result contains values that do not exist in the business materials |
| Tool calling node | The model decides whether to call based on the tool description, parameter descriptions, and whether required, and may fabricate parameter values when there are no required parameters | Check the input parameters and return in the call record to confirm whether it is a follow-up or a call according to the description; check toolChoice, functionCall, and the connection status of the tool calling node |
Three common errors
- Symptom: the name, quantity, or date of the same record does not match between the review invitation and the expert review comments. Cause: the two materials arrive in separate batches, and the fields enter the tool calling node without cross-comparison.
- Symptom: the versions of the author revision notes and the expert review comments are not synchronized, and response items and comment items are misaligned. Cause: the version identifier is not updated with the materials, and the tool return result continues to use the old version's values.
- Symptom: the review invitation lacks a signature or receipt, yet the tool call still returns a result. Cause: required parameters are not guided to be asked as follow-ups in the prompt, and the model calls directly according to the tool description.
How to verify the setup
Acceptance of this feature: test-run with a question missing a required parameter, check the input parameters and return in the call record, and confirm whether it is a follow-up or a call according to the description; check toolChoice, functionCall, and the connection status of the tool calling node.
Business test: ask a question missing key information and see whether it asks a follow-up before answering.
- Open the tool node description fields, check whether this description can be understood on its own, and confirm it states the purpose, when to use it, and the return result.
- Open each parameter configuration one by one, check the two items of parameter description and whether it is required, and test-run with a question missing that field to see whether it asks a follow-up first instead of fabricating directly.