Custom

Plain-language policy defined by your prompt string.

YAML key: custom
Direction: input

Plain-language policy defined by your prompt string.

Configuration

Prop

Type

defend.config.yaml (fragment)

guards:
  input:
    modules:
      - custom:
          prompt: "Stay within policy."

Configure under guards.input.modules when using an LLM-backed input provider.

Ban competitors output

Competitor restrictions on model output.

Custom output

Plain-language output policy defined by your prompt string.