Create policy

Decide which tools are allowed, which need a human, and which models can run. Advanced limits stay out of the way.

Tools shown below are collected from all registered agents in your workspace. A policy's scope (agent selector + environment) controls which agents it applies to — only set access for tools used by the agents this policy will govern.

Policy name

Which agents it governs is set by assigning the policy to an agent

Policy nameStable name for audit records and the policy list.

01 · Pre-flight

Can this agent run? Checked before execution begins.

Tools

0 allow · 0 review · 5 deny

Everything not marked Allow or Review is denied by default. Review tools pause for a human before they run.

Tool

Allow

Review

Deny

Remove

delete_research_data

Allow

Review

Deny

draft_outreach_email

Allow

Review

Deny

read_customer_profile

Allow

Review

Deny

score_icp_fit

Allow

Review

Deny

send_outreach_email

Allow

Review

Deny

Add tool name

Models

Which LLMs this policy permits

Allowed models

All 7 models allowed▼

claude-fable-5claude-opus-4-8claude-sonnet-5gemini-3.5-flashgpt-4ogpt-4o-minigpt-5-mini

Select which models this policy permits. Leaving all unchecked denies every model request.

02 · In-flight

Is this action safe and worth it? Checked on every tool call.

Guardrails

Prompt and payload checks on every governed call

When we see PII

Prompt injection

Data leaving the org

03 · Post-flight

Can we explain and improve it? Tracks behavior and feeds back into pre-flight.

Privilege de-escalation

Auto-tighten access when an agent misbehaves

De-escalation threshold (strikes)After this many token-overage strikes, all tools require human review until an operator resets the agent.

Cancel