App takedown & content moderation
Why apps are sometimes taken down
Emergent runs automated content-policy and phishing screening on all published apps to keep the platform safe and trustworthy for everyone. If an app triggers these filters - even by mistake - it may be automatically taken offline.
False positives happen
Automated moderation occasionally flags legitimate apps. We provide a simple dispute and recovery path so you're never permanently locked out of your work.
Common triggers include:
- Phishing keywords or suspicious login flows (e.g. credential harvesting forms)
- Malicious content or content-policy violations (harassment, spam, illegal material)
- Impersonation of well-known brands or services
- Misleading or deceptive UI patterns
What happens when your app is flagged
- Immediate takedown: The live published app is removed from public access; the public URL returns a
or takedown notice.403 Forbidden - Notification: You receive an email and in-workspace alert explaining the reason for takedown.
- Source code remains safe: Your workspace, chat history, and code are not deleted - only the public published app is affected.
Warning
Your credits are not refunded for apps taken down after published app. The published app consumed platform resources (build, hosting), so the charge stands.
How to dispute or fix the issue
Review the takedown notice
Check your email and workspace notifications for the specific policy violation cited. Many false positives stem from keywords in placeholder text, demo content, or authentication flows.
Submit a dispute (false positive)
If you believe your app was flagged in error, click Dispute takedown in the workspace banner or reply to the notification email. Include:
- A brief description of the app's purpose
- Why the flagged content is legitimate (e.g. "auth form is for a real user login, not phishing")
Our support team reviews disputes within 24 hours on business days. If approved, your app is restored to its original public URL.
Fix and re-publish (genuine violation)
If the takedown was correct - or if you prefer to fix the issue rather than wait for review - edit your app in the workspace:
- Ask the agent to remove or reword the problematic content
- Test the changes in preview
- Click Re-publish to publish a new version
The new published app will be re-screened automatically. If it passes, it goes live immediately.
Fork and replace (persistent issues)
If a specific workspace or app keeps triggering false positives, you can fork the project:
- Export your code (via Github push in the workspace)
- Start a new workspace and upload or describe the app again
- Publish from the clean slate
The new published app gets a fresh URL and a new screening pass.
Preventing false positives
Avoid phishing keywords
Words like "verify your account," "urgent action required," or fake login prompts trigger phishing filters. Use clear, honest language in forms and CTAs.
Label demo content clearly
If your app includes placeholder or test data that resembles policy violations (e.g. fake reviews, mock scams for educational purposes), add visible labels like "Demo only" or "Example content".
Use real branding sparingly
Avoid using trademarked logos, brand names, or UI that mimics well-known services unless you have permission or a clear parody/educational use case.
Review before publishing
Preview your app in the workspace before publishing. Look for anything a moderation system might misinterpret as malicious.
Appeals and repeat offenders
- First takedown: Dispute or fix freely; no penalty.
- Multiple violations: Repeated policy breaches (especially after disputes are rejected) may result in account suspension or permanent bans.
- Malicious intent: Apps designed to harm users, steal credentials, or violate laws result in immediate account termination with no refund.
Tip
Emergent's moderation is designed to be lenient for edge cases. If you're building something unconventional (satire, security research, etc.), reach out to support before publishing so we can whitelist or review it manually.
