Permission Request
Permintaan Izin
What the harness shows the user before executing a tool call that is not pre-approved.
A permission request is the prompt a harness shows before it executes a tool call that is not already pre-approved. It is the mechanism that puts a human in the loop: the agent asks, the person decides, the action proceeds or does not.
The prompt has to carry the information needed to decide. Run rm? is not enough. Delete 240 files in src/generated, not recoverable is enough. A permission request that does not say what will happen cannot be approved meaningfully, and one that fires on every safe action trains the reader to click through all of them, including the one that mattered.
This is where the harness, not the model, does the work. The model proposes; the harness checks the policy and asks. That separation is what makes the guarantee real, because a rule written inside a system prompt is only a suggestion to a system optimised to be helpful.
Related terms
Permission Mode
The permission-gating slice of an agent mode: which tool calls trigger a request and which run automatically.
Human in the Loop
A person who reviews, approves, or corrects the agent's work at key points.
Guardrails
Checks that stop or constrain an action before it happens.
Approval
The moment a person says yes before the agent does something it cannot undo.
Tool Call
The model's request naming a tool and its arguments. Structured output, not a side effect.
Want this working inside your team?
It starts with one conversation about your workflow.