Edit policy

Change the live rules. Publishing creates a new version and leaves activation to you on the next screen — or activate from History when ready.

01 · Pre-flight

Can this agent run? Checked before execution begins.

Tools

2 allow · 1 review · 0 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

arxiv_search

Allow

Review

Deny

delete_research_data

Allow

Review

Deny

web_search

Allow

Review

Deny

Add tool name

Which LLMs this policy permits

Allowed models

1 of 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.

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.

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.