NDA Terms Validation

NDA review involves checking a set of proposed terms against a fixed policy — minimum confidentiality periods, approved governing jurisdictions, required clauses, and liability thresholds. Done manually, this is repetitive and error-prone. Different reviewers apply the standards inconsistently, and corrections are written in free text rather than structured data. This template automates that process — the model applies the same company standards on every call and returns a structured result your legal workflow can act on directly.

The agent receives structured NDA metadata and a set of proposed terms, and validates them against the company's minimum NDA standards defined in the attached policy document. It checks each field independently and returns a validation summary with pass/fail counts, a corrected terms object containing the compliant value for every field, and a final verdict.

If a field passes, its original submitted value is returned unchanged. If a field fails, the corrected minimum-compliant value is returned in its place — ready to send back to the counterparty without manual intervention. If a field cannot be evaluated because the input data is insufficient, it returns null rather than a guess. Any downstream rule receiving a null can detect it explicitly and route the record to manual review rather than processing an incomplete result silently.

Problem: All corrected_terms fields return null.
Solution: This is expected when the submitted terms object is missing or empty. The rule is designed to return null rather than estimate — check that all term fields are populated before resubmitting.

Problem: Individual corrected_terms fields return null while others are populated.
Solution: A null on a specific field means its correct value could not be determined from the input. For example, non_compete_duration_months returns null when non_compete_included is false, because the field is not applicable. Check whether the field is conditional on another field's value and ensure the dependency is present.

Problem: The rule cannot be executed and shows a warning on the Attachments tab.
Solution: The selected AI model does not support file input. Either switch to a model that accepts attachments or remove the attachment and embed the NDA standards as plain text directly in the prompt.