- SAFE
- UNSAFE
- TAGGED
- MANUAL
Short answer
How can my AI moderate images?
Your AI uses FileStack to check if uploaded files are safe for work and to identify specific objects within them. This allows you to automatically flag problematic uploads and organize your library by visual content.
Content outcomes
Where your files land.
The AI sorts every upload into one of these four categories.
SAFE
Approved content
Files that pass safety checks move directly to your storage. Your AI confirms they are clean for public viewing.
UNSAFE
Flagged for review
The AI identifies inappropriate content and marks it as unsafe. You get a clear signal to prevent these files from appearing.
TAGGED
Organized by objects
Images get labeled with specific visual tags. This makes your files searchable based on what is actually in them.
MANUAL
Human intervention
If the AI is unsure about a file's safety, it flags it for you. You decide whether to approve or reject these edge cases.
The workflow
What your AI does when an image arrives.
The AI handles the heavy lifting of visual inspection so you don't have to look at every file.
Check safety
The AI immediately looks at the file to see if it violates safety guidelines.
get_sfw_statusIdentify objects
Once cleared, the AI scans the image to see what objects are present.
get_image_tagsApply labels
The AI uses the identified objects to create descriptive tags for your database.
get_image_tagsFlag violations
If the safety check fails, the AI marks the file as unsafe immediately.
get_sfw_status
Try it
Copy these to start.
Use these prompts to begin moderating your files.
Starting points
Replace the bracketed text with your specific folder names or channel IDs.
Filestack Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_KGFK0vBlAyeq8lyImuHLANnNzsFyop3aMeVzgPza/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Filestack capabilities are ready to use.
{
"mcpServers": {
"filestack-mcp": {
"url": "https://edge.vinkius.com/vk_preview_KGFK0vBlAyeq8lyImuHLANnNzsFyop3aMeVzgPza/mcp"
}
}
}Scan the latest uploads in my 'user-images' folder and tell me if any are unsafe.
Look at the images in my 'product-catalog' channel and list all the objects you find.
Check the safety status of the files in my 'uploads' directory and flag anything problematic.
Tag all the images in my 'social-media' folder based on the objects they contain.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect FileStack once, then ask.
Just link your account once. Your credentials stay encrypted, and you can start asking your agent to scan files immediately.
Connect FileStack to your AIFAQ
How this task behaves.
- 01
Can the AI delete unsafe images automatically?
No. The AI can identify and flag unsafe content, but you must perform the actual deletion yourself.
- 02
Does the AI change the actual image files?
No. The AI only reads the files to check for safety and to extract tags. It does not modify the image data.
- 03
Can it identify text inside images?
No. This specific setup is designed for safety status and object recognition, not OCR.
- 04
How does it know if an image is safe?
It uses the safety status tool to check the file against predefined safety standards.
- 05
Can I use this to organize my photo library?
Yes. By using the object tagging capability, your AI can categorize images based on what it sees.
More questions about FileStack? The Connector page answers them. See everything the FileStack Connector can do
Connect Filestack to Claude, Cursor, ChatGPT & more
