Defend
ReferenceModules

Tool misuse

Constrain tool-call abuse in model output.

View on GitHub

YAML key: tool_misuse
Direction: output

Constrain tool-call abuse in model output.

Configuration

Prop

Type

Example

defend.config.yaml (fragment)
guards:
  output:
    enabled: true
    provider: claude
    modules:
      - tool_misuse:
          allowed_tools: []
          max_calls_per_turn: 3

Provider usage

Configure under guards.output.modules with output guarding enabled and provider claude or openai.

See the modules overview.

Code execution output

Flag dangerous code execution patterns in model output.

Excessive agency

Detect overreach beyond permitted autonomy.

On this page

ConfigurationExampleProvider usage