VerifEye — Claude Code MCP server: face recognition and comparison
An MCP server that analyzes faces in photos: estimates age and gender, detects emotions and attention, onboards faces into collections, and compares images via VerifEye Face API.
no license ★ 1 updated 2026-10-16
claude mcp add --transport http verifeye https://verifeye-mcp-server.realeyes.ai/mcpWhat VerifEye does
The server runs remotely: connect the address https://verifeye-mcp-server.realeyes.ai/mcp in your MCP client, sign in through the browser via OAuth, and the agent gains face analysis tools. verifeye_identify_face estimates apparent age, gender, emotions and attention from an image URL. verifeye_onboard_face stores a face template in a collection to identify the same person later. verifeye_compare_face matches a photo against one onboarded face.
Collections are per region — EU and US. Calls default to EU unless you specify otherwise. Images are passed as public JPEG or PNG URLs, which the server fetches itself. You need a VerifEye Developer Console account; the API key stays server-side.
Hosted by Realeyes, no local installation required. verifeye_delete_face and verifeye_delete_collection destroy data irreversibly.
What it helps with
- estimate apparent age and gender from a photograph
- store a face in a collection for later identification
- check whether a face from a photo already exists in a collection
- compare two face images and get a match verdict
- delete a face template or wipe an entire collection
Example prompts for the agent
- Estimate the age and gender of the face in the photo at this URL
- Create a collection called
front-deskand onboard the face from this image with keyvisitor-1042 - Check if the person in
today.jpgis already in myfront-deskcollection
How to install VerifEye
Platform CLI: remote server
claude mcp add --transport http verifeye https://verifeye-mcp-server.realeyes.ai/mcpPlatform CLI: remote server
codex mcp add verifeye --url https://verifeye-mcp-server.realeyes.ai/mcpConfig file
{
"mcpServers": {
"verifeye": {
"url": "https://verifeye-mcp-server.realeyes.ai/mcp"
}
}
}Config file
{
"servers": {
"verifeye": {
"type": "http",
"url": "https://verifeye-mcp-server.realeyes.ai/mcp"
}
}
}Platform CLI: remote server
gemini mcp add -t http verifeye https://verifeye-mcp-server.realeyes.ai/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about VerifEye
How to install VerifEye in Claude Code?
Run claude mcp add --transport http verifeye https://verifeye-mcp-server.realeyes.ai/mcp, then sign in via OAuth in the browser when prompted.
What does VerifEye do?
Recognizes faces in photos through MCP tools: estimates age, gender, emotions and attention, stores face templates in collections, compares images and identifies matches.
How does VerifEye receive the image?
Tools accept a public JPEG or PNG URL, which the server fetches itself. Images attached to the conversation cannot be passed to tools: upload the image and give the agent that link, preferably a short-lived pre-signed one.