- CLEAN
- FLAGGED
- REVIEW
- LOGGED
Short answer
How can I use AI to moderate text and images?
You can use your AI to scan incoming messages and uploaded files for profanity or NSFW content. The agent uses JigsawStack to flag violations and categorize the results. You get a clear breakdown of what needs attention.
Content outcomes
Where every flag lands.
The AI sorts every piece of data based on what it finds.
CLEAN
Safe content
The AI finds no profanity or NSFW material. These items pass through without any further action.
FLAGGED
Violations detected
The agent identifies profanity or inappropriate images. You receive a report detailing the specific issue found.
REVIEW
Needs a human
The AI flags borderline cases that require a person to make the final call. This keeps your moderation accurate.
LOGGED
Audit trail
Every scan result is ready for your records. You can review the history of what was checked and what was caught.
The workflow
What your AI does when content arrives.
The AI acts as a first line of defense for your digital spaces.
Text screening
Your agent looks at text strings to find offensive language or profanity.
check_profanity_textImage screening
The AI inspects uploaded images for NSFW content.
check_nsfw_imageClassification
The agent categorizes the severity of the detected content.
check_profanity_textcheck_nsfw_imageReporting
The AI summarizes the findings so you can act on them immediately.
check_profanity_text
Try it
Copy these to start.
Use these prompts to trigger the moderation tools.
Starting points
These are basic ways to start. Swap out the channel names or file paths for your actual data.
JigsawStack Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_KPxx7QRVy9rKRYnDNTRkP3T5sok2axbozjs2aHFw/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 — JigsawStack capabilities are ready to use.
{
"mcpServers": {
"jigsawstack-mcp": {
"url": "https://edge.vinkius.com/vk_preview_KPxx7QRVy9rKRYnDNTRkP3T5sok2axbozjs2aHFw/mcp"
}
}
}Check this Slack channel for any profanity in the last ten messages.
Scan the images in this Google Drive folder for NSFW content.
Look through the latest comments in this GitHub repo and flag any profanity.
Review the images I just uploaded to this folder and tell me if any are NSFW.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect JigsawStack once, then ask.
You only need to link the MCP once. Your credentials stay encrypted, and you can start moderating immediately.
Connect JigsawStack to your AIFAQ
How this task behaves.
- 01
Can the AI automatically delete bad content?
No. The AI identifies and flags content, but you must perform the actual deletion yourself.
- 02
Does the AI moderate content in real time?
Yes, once you give the command to scan a specific source, the AI processes the data immediately.
- 03
Can it moderate video files?
No. The current tools are limited to text and images.
- 04
Will the AI change the original message?
No. The AI only reads and analyzes the content; it does not edit or rewrite the original text or images.
- 05
How does it handle slang or coded language?
The AI uses JigsawStack to detect profanity, which includes common variations, but it may miss highly specific or evolving slang.
More questions about JigsawStack? The Connector page answers them. See everything the JigsawStack Connector can do
Connect JigsawStack to Claude, Cursor, ChatGPT & more
