# Yitu Technology MCP for AI Agents AI Agent Connect > Yitu Technology is an enterprise computer vision platform for facial recognition and identity verification. It lets your AI agent manage facial repositories, detect faces in images, and perform high-precision 1:1 and 1:N matching. ## Overview - **Category:** artificial-intelligence - **Price:** Free - **Endpoint:** https://edge.vinkius.com/undefined/ai-agent-connect - **Tags:** computer-vision, facial-recognition, biometric-authentication, identity-verification, image-analysis, security-surveillance ## Description Yitu Technology lets you put enterprise-grade facial recognition into your AI agent's hands. If you're tired of clicking through complex security dashboards or manually cross-referencing identity files, this is the fix. You can tell your agent to find specific people in image URLs, verify if two photos show the same person, or search through private facial databases without ever touching a technical console. It handles the heavy lifting of identity indexing and repository management so you can focus on the actual work. Whether you're managing a high-volume archive of digital identities or setting up an automated security checkpoint, you get a direct line to high-precision computer vision. You'll stop wrestling with data entry and start getting answers in a natural conversation. By connecting this to your workflow via Vinkius, you get a reliable way to manage visual intelligence without the typical overhead of building your own infrastructure. This means you can quickly audit access logs, register new identities in bulk, or perform liveness checks to spot spoofing attempts. It turns your AI into a coordinator for your visual data, making it easier to manage large-scale biometric systems from a single chat interface. ## Tools ### detect_active_liveness Returns per-action pass/fail. This confirms a person is real by having them perform a specific action. ### add_face_to_repo Register a face in a repository. Use this to index new identities for future searches. ### compare_faces Verify if two faces match. This is useful for 1:1 identity verification between two images. ### create_face_repo Create a new facial repository. Use this to organize your facial data into distinct collections. ### delete_face_repo Delete a facial repository. This helps you clean up old or unnecessary identity archives. ### detect_face Detect faces in an image. This allows your agent to find and locate people within any provided image URL. ### search_face_in_repo Search for a face in a repository. This performs 1:N recognition to find matches within a private database. ### list_repos List all facial repositories. Use this to see every active collection you have in your system. ### moderate_image Content moderation for images. This checks your visual data to ensure it meets your safety standards. ### ocr_id_card Extract text from an ID card image. This turns physical ID photos into usable text data automatically. ### remove_face_from_repo Remove a face from a repository. This is the standard way to delete a specific person from a collection. ### detect_silent_liveness Detects photos, screens, and 3D masks. Use this to prevent people from using photos to trick your security. ## Prompt Examples **Prompt:** ``` Find everyone in this photo: [URL] ``` **Response:** ``` I've analyzed the image using Yitu Technology. I detected 3 unique faces in the photo. Would you like me to extract the facial locations or perform a comparison against your existing repositories? ``` **Prompt:** ``` Is this person in our 'Staff_List' repo? ``` **Response:** ``` I've performed a search in repository `Staff_List`. I found a match with 98.2% confidence for Person ID `staff_042`. Should I retrieve the full metadata associated with this person? ``` **Prompt:** ``` Check if this ID card is real. ``` **Response:** ``` I've processed the ID card image. Here is the extracted information: | Field | Value | | :--- | :--- | | Name | John Doe | | ID Number | 123456789 | | Expiry | 2030-01-01 | I can also perform a liveness check if you'd like to verify the person is present. ``` ## Capabilities ### Verify identities Check if two faces match with high precision to confirm a person's identity. ### Search facial databases Find matches for a face within your private collections using 1:N recognition. ### Manage face repositories Create, list, and delete facial data collections to organize your identity archives. ### Detect faces in images Locate faces within any image URL and extract their specific locations. ### Extract ID card text Pull text from ID card images automatically using OCR capabilities. ### Spot spoofing attempts Detect photos, screens, or 3D masks to ensure you are interacting with a real person. ## Use Cases ### Security Checkpoint Verification A security guard asks the agent to verify a person's face against a staff list using compare_faces to confirm they are authorized to enter. ### Automated ID Data Entry A user uploads a photo of their ID, and the agent extracts the text using ocr_id_card to populate a registration form. ### Large-Scale Archive Search An admin asks the agent to find all matches for a specific face in a private collection using search_face_in_repo. ### Anti-Spoofing for Remote Access A remote login system uses detect_active_liveness to ensure a real person is performing an action rather than holding up a tablet. ## Benefits - Speed up identity verification by using compare_faces to match individuals instantly. - Automate archive updates with add_face_to_repo to keep your records current without manual entry. - Spot spoofing attempts with detect_silent_liveness to block photos or 3D masks from your security gate. - Extract data from IDs with ocr_id_card to turn physical documents into digital text immediately. - Keep content safe with moderate_image to ensure all uploaded visual data meets your safety standards. - Manage many collections easily with list_repos to keep your various facial databases organized. ## How It Works The bottom line is you get enterprise facial recognition capabilities inside your favorite AI chat interface. 1. Subscribe to this Connector on the Vinkius Marketplace. 2. Enter your Yitu AppID (DevId) and APIKey (DevKey) into your AI client settings. 3. Start managing your visual identity infrastructure through Claude, Cursor, or any other MCP-compatible client. ## Frequently Asked Questions **Can Yitu Technology MCP help with enterprise security?** Yes. It provides enterprise-grade facial recognition and liveness detection to help secure access points and audit identities. **How does Yitu Technology MCP handle spoofing?** It uses specific tools to detect if a person is holding up a photo, a screen, or a 3D mask instead of being physically present. **Can I use Yitu Technology MCP to manage my own database?** You can manage private facial repositories, including creating new ones, listing existing collections, and adding or removing faces. **Does Yitu Technology MCP work with Claude?** Yes, it works with Claude and any other MCP-compatible client like Cursor or VS Code. **How accurate is Yitu Technology MCP for face matching?** It provides high-precision 1:1 face comparison and 1:N recognition for matching identities within your private repositories. **Can Yitu Technology MCP extract text from IDs?** Yes, it includes OCR capabilities specifically designed to pull text from ID card images. **How do I find my Yitu AppID and APIKey?** Log in to the [Yitu Cloud Platform](https://www.yitutech.com/), navigate to the 'Developer Center' or 'API Management' section to find your unique AppID (DevId) and APIKey (DevKey). **What is a facial repository?** A facial repository is a secure, private database where you store facial features (templates) of known individuals. This allows the system to perform 1:N searches to identify a person from a large group. **How accurate is the identity comparison?** Yitu is a world leader in facial recognition accuracy. The system returns a confidence score (typically 0.0 to 1.0). A score above 0.8 is generally considered a highly reliable match for the same person.