Privent icon

Privent

Privent DLP: session-scoped tokenization, detokenization, risk scoring, audit events and agent handoffs for AI agent workflows.

Actions6

Risk Check → Score

AI-generated

Summary

Scores the text in each input item with Privent’s risk-scoring service and returns the score with the item. Use this operation with API Key or Tokenless authentication; Local authentication does not expose its input fields in the UI.

Inputs

  • Authentication — Choose API Key for full Privent access, or Tokenless for visitor-based risk scoring when the backend has visitor auth enabled. Although Local is an available authentication option, the Score operation’s fields are not shown for it.
  • Text Field (required) — Name of the input item property containing the text to score; defaults to text.
  • Trace ID — Correlation ID from an upstream Privent Session node; a fresh ID is generated if none is supplied.
  • Agent Name — Optional logical agent identifier from an upstream Privent Session node, recorded on the audit event.

Output shape

A list of scored items, corresponding to the input items.

Each item includes the risk score result. The operation groups scoring requests into a batch when possible; on a batch-length mismatch it retries items individually and attaches any per-item failures as errors.

Links

Discussion