dry_run
An API design pattern that allows callers to preview the execution result without actually performing the operation, commonly used for pre-validation of high-risk operations and in the design of agent safety mechanisms.
Timeline (last 90 days)
In Bot skill engineering, the default value of dry_run should be True, making the system take the simulation path by default.
For high-risk operations, a dry_run flag should be provided, letting the Agent first show 'what will happen' without actually executing it.
Setting the dry_run default to True is a secure-by-default design pattern, and developers must explicitly turn off simulation to trigger real operations
All Facts (3)
In Bot skill engineering, the default value of dry_run should be True, making the system take the simulation path by default.
50%UnverifiedFor high-risk operations, a dry_run flag should be provided, letting the Agent first show 'what will happen' without actually executing it.
50%UnverifiedSetting the dry_run default to True is a secure-by-default design pattern, and developers must explicitly turn off simulation to trigger real operations
50%