- TAGS
- FIELDS
- SEARCH
- HUMAN
Short answer
How can I use AI to update my Canto metadata?
Your AI uses the Canto MCP to pull existing technical properties and then writes new values directly into your custom fields or tags. You get a clean, searchable library without manual entry. Here is how the process works.
Metadata outcomes.
Where your data lands.
What happens after your AI processes your assets.
TAGS
Updated tag lists
The AI rewrites specific tags based on what it sees in the image data. Your search results become much more accurate.
FIELDS
Custom field updates
The agent pushes new values into your Canto custom fields. This keeps technical specs in sync with the actual files.
SEARCH
Better asset discovery
Because the metadata is accurate, you find what you need instantly. No more scrolling through thousands of untagged files.
HUMAN
Final review needed
The AI performs the writes, but you decide which assets need changes. You maintain control over what actually gets updated.
The workflow
What your AI does when the request arrives.
The AI follows a simple loop of reading data and applying changes.
Read properties
The agent pulls the current technical details and existing tags from your Canto assets.
get_image_metadataAnalyze data
Your AI interprets the technical properties to determine what the new tags or field values should be.
get_image_metadataApply changes
The agent sends the new metadata values back to Canto to overwrite the old information.
patch_image_metadataVerify updates
The AI checks the updated fields to ensure the new tags were applied correctly.
get_image_metadatapatch_image_metadata
Try it
Copy these to start.
Use these prompts to trigger the Canto MCP.
Starting points
These are basic prompts. Swap out the specific field names or asset IDs to match your Canto setup.
Canto Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_k5M8Qnn3Ob8bW7XWthumvaAxvpjn4zlYc14TkuIE/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 — Canto capabilities are ready to use.
{
"mcpServers": {
"canto-mcp": {
"url": "https://edge.vinkius.com/vk_preview_k5M8Qnn3Ob8bW7XWthumvaAxvpjn4zlYc14TkuIE/mcp"
}
}
}Read the metadata for asset ID 12345 and tell me what the current tags are.
Update the 'Color' custom field for all images in the 'Summer Campaign' folder to 'Blue'.
Look at the technical properties of my latest upload and add 'High-Res' to the tags.
Find all assets with the tag 'Draft' and change their status field to 'Approved'.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Canto once, then ask.
Just link your Canto account to Vinkius. Your credentials stay encrypted, and you can immediately start talking to your assets through any compatible client.
Connect Canto to your AIFAQ
How this task behaves.
- 01
Can the AI delete images from my Canto library?
No. The tools only allow for reading and patching metadata. It cannot delete files.
- 02
Will the AI change my folder structure?
No. This MCP is strictly for updating metadata and tags, not for moving or deleting folders.
- 03
Does the AI need to see the actual image to tag it?
The AI reads the technical metadata provided by the tool. It uses those properties to decide how to rewrite the tags.
- 04
Can I review changes before they happen?
Yes. You can ask your AI to show you what it plans to change before you tell it to apply the updates.
- 05
Can the AI update metadata for every asset at once?
The AI can process assets one by one or in small batches as you direct it through your chat interface.
More questions about Canto? See everything the Canto Connector can do
Connect Canto to Claude, Cursor, ChatGPT & more
