allow words and rules

allowed words and rules tell the detector to stop flagging something that is valid for the user's voice or domain.

updated may 3, 2026 voice check no images
quick answer

use allow actions when a repeated flag is clearly wrong.

use this for

  • reducing false positives
  • protecting brand-specific language
  • keeping technical terms from being flagged

steps

  1. open a tooltip on a false positive.
  2. choose allow word for a specific term.
  3. choose allow rule when the whole rule is wrong for this voice.
  4. run the check again.
  5. review fewer future false positives.

details to know

  • allowed words and rules live on the profile.
  • they affect future ai-pattern checks.
  • telemetry records allow actions so product can see noisy rules.

limits and edge cases

  • do not allow a rule because one sentence is inconvenient.
  • too many broad allowances can weaken the detector.
ready to use the product instead of reading about it? open app