Skip to main content

Submitting an Agent for Review

Agent owners submit their agents for agent review by completing a security questionnaire. This page walks through the entire submission process from start to finish.

Key Concepts

Before you start

  • Your agent must be in draft status.
  • The Agent Review must be enabled on your organization.
  • You must be the owner of the agent (or an org admin).
  • Your agent's environment (non-prod or prod) must match agent review's required environment setting.

What the submission includes

A submission contains your answers to the security questionnaire (covering data flow, tool scope, autonomy, network access, compliance, privacy, and prompt security), plus your selections for which MCP servers and skills the agent should have access to. These answers are compiled into a guardrail profile that enforces security rules at runtime.

Step-by-Step Guide
1

Open your agent's detail page

Navigate to AI Company → Agents and click on the agent you want to submit. Or go directly to /dashboard/ai-company/agents/[id].

Agent detail page showing the agent name, description, model, and a 'Submit for Review' button
The agent detail page shows a 'Submit for Review' button when agent review is enabled and the agent is in draft status.
2

Complete identity verification

Before the questionnaire opens, you'll need to verify your identity. This prevents automated or accidental submissions.

  • If you have SignedApproval connected: A push notification is sent to your mobile app. Approve it within 5 minutes.
  • Otherwise: Complete a Cloudflare Turnstile CAPTCHA challenge.
Pre-verification step showing either the SignedApproval push approval waiting screen or a Cloudflare Turnstile CAPTCHA
Identity verification ensures submissions are intentional. SignedApproval users see a push notification; others complete a CAPTCHA.
3

Select the deployment environment

The first step of the wizard asks which environment this agent will run in:

  • Non-Production — Development, staging, or testing environments.
  • Production — Live environments serving real users or data.
  • Both — The agent will run in both environments.
4

Answer the security questionnaire

Work through the multi-step wizard. The questionnaire has 14 steps covering:

  1. Environment — Deployment target
  2. Operations — Expected workload description
  3. Triggers — How the agent is activated
  4. Data Flow (D1–D7) — PII, PHI, financial data, persistence, key interception
  5. Tool Scope (T2–T5) — Browser automation, models, APIs, write permissions
  6. Autonomy (A1–A6) — Budget, rate limits, session duration, approval gates
  7. Network (N1–N3) — Domain restrictions, private networks, URL blocking
  8. Compliance (C1–C6) — Frameworks, audit logging, retention
  9. Privacy (PV1–PV7) — Data storage, third-party sharing, minors, cross-border
  10. Prompt Security (P1–P4) — Untrusted input, multi-step chains, system prompt protection
  11. Security Context (S1–S6) — Authorized sources, exfiltration boundaries, threat model
  12. Review — Summary of all answers
  13. Key Access — Provider key bindings for the agent
  14. Verify & Submit — Final confirmation
Review submission wizard showing the Data Flow step with yes/no toggles for PII, PHI, and financial data questions
Each step presents relevant questions with tooltips explaining what each setting controls. Hover over the info icon for details.
Tip
Each question has a tooltip icon explaining its purpose and impact. Hover over it to understand how your answer affects the guardrail profile. When in doubt, choose the more restrictive option — you can always resubmit with looser settings later.
5

Select MCP servers and skills

During the Tool Scope section, you'll select which MCP servers and skills the agent should have access to. Only governed integrations will be available at runtime — any tool not on this list will be blocked by the gateway.

6

Review and submit

The Review step shows a summary of all your answers. Check everything carefully — once submitted, you can only withdraw and resubmit, not edit in place. Click Submit for Review to send the submission to reviewers.

Review step of the submission wizard showing a summary of all answers organized by category, with a Submit button at the bottom
The review step shows every answer before submission. Double-check everything before submitting.
7

Wait for agent review

Your submission is now pending. Reviewers will be notified (if they have notifications enabled). You can track the status on Agent Review page or in your agent's detail view.

If your submission is rejected, you'll receive feedback in the rejection rationale. You can address the concerns and resubmit.

Important
If your agent is governed and you later change its model, system prompt, expertise, or environment, the governance is invalidated. You'll need to resubmit for review. See Drift Detection for details.