mcpField
T
Tencent Ekyc Mcp Server

Tencent Ekyc Mcp Server

Local
github.com/FrogOVOSecurity0Updated 27w ago

About Tencent Ekyc Mcp Server

Tencent Cloud eKYC MCP Server wraps Tencent Cloud eKYC APIs as tools for AI agents handling identity checks and KYC workflows. It covers face matching, liveness, deepfake detection, and asynchronous result queries.

Key features

  1. 01Match faces with ID photos
  2. 02Run silent liveness checks
  3. 03Detect deepfake face images
  4. 04Check image quality before submission
  5. 05List supported document types by region
  6. 06Track asynchronous verification results

Use cases

  1. 01Verify identity with an ID photo and selfie video
  2. 02Screen AI-generated or altered face images
  3. 03Validate image quality before KYC submission
  4. 04Retrieve results of pending checks

MCP support

This MCP server exposes Tencent Cloud eKYC API capabilities as tools for MCP clients, including identity verification, liveness, deepfake detection, and status queries.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"tencent-ekyc": {
"command": "npx",
"args": [
"-y",
"tencent-ekyc-mcp-prototype"
],
"env": {
"TENCENTCLOUD_SECRET_ID": "<TENCENTCLOUD_SECRET_ID>",
"TENCENTCLOUD_SECRET_KEY": "<TENCENTCLOUD_SECRET_KEY>"
}
}
}
}
Run commandnpx -y tencent-ekyc-mcp-prototype
Environment variables to set
TENCENTCLOUD_SECRET_IDSecret
TENCENTCLOUD_SECRET_KEYSecret

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmtencent-ekyc-mcp-prototype

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Tencent Cloud eKYC MCP Server

A Model Context Protocol (MCP) server that wraps Tencent Cloud eKYC APIs as callable tools for AI agents (Claude Desktop, Cursor, Windsurf, etc.). Enables face verification, liveness detection, deepfake detection, and identity verification — all via natural language.

Features

  • Face Verification (1:1) — Compare ID card photo with selfie video
  • Liveness Detection — Anti-spoofing with silent challenge
  • Deepfake Detection — Detect AI-generated or manipulated face images
  • Image Quality Check — Pre-verify image quality before KYC submission
  • Supported Documents Query — List all supported ID types by region
  • Verification Status Tracking — Check async verification results

Supported Regions & Documents

RegionDocuments
Indonesia (ID)KTP (ID Card), SIM (Driver License)
Malaysia (MY)MyKad, MyTentera
Thailand (TH)Thai ID Card
Singapore (SG)NRIC
Philippines (PH)UMID, SSS ID
Vietnam (VN)Chip-based ID
More...16 regions, 28 document types

Quick Start

Prerequisites
  • Node.js >= 20
  • Tencent Cloud account with eKYC service enabled
  • SecretId & SecretKey from Tencent Cloud Console
Install & Build
git clone https://github.com/FrogOVO/tencent-ekyc-mcp-server.git
cd tencent-ekyc-mcp-server
npm install
npm run build
Configure Claude Desktop

Edit %APPDATA%\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "tencent-ekyc": {
      "command": "node",
      "args": ["D:\\path\\to\\tencent-ekyc-mcp-server\\dist\\index.js"],
      "env": {
        "TENCENT_SECRET_ID": "your-secret-id",
        "TENCENT_SECRET_KEY": "your-secret-key",
        "TENCENT_REGION": "ap-singapore",
        "EKYC_LOG_LEVEL": "info"
      }
    }
  }
}

Restart Claude Desktop. You should see tencent-ekyc tools available.

Usage (Natural Language)

Once configured, just talk to your AI agent:

Please verify this person's identity using their KTP photo and selfie video.

The agent will automatically call tencent_ekyc_verify_identity with the correct parameters.

Available Tools

Tool NameDescription
tencent_ekyc_get_credential_statusCheck if API credentials are configured
tencent_ekyc_get_supported_documentsList supported ID documents by region
tencent_ekyc_check_image_qualityValidate image quality before submission
tencent_ekyc_verify_identityFull KYC: face comparison + liveness + verification
tencent_ekyc_detect_deepfakeDetect deepfake/AI-generated face images
tencent_ekyc_get_verification_statusQuery async verification result

Project Structure

src/
├── index.ts                    # MCP server entry point
├── auth/tc3-signer.ts         # TC3-HMAC-SHA256 signature
├── config/
│   ├── documents.ts            # Supported document matrix
│   └── env.ts                 # Environment variable validation
├── enhancer/
│   ├── error-mapper.ts         # Error code → AgentHint mapping
│   └── response-enhancer.ts   # Response enrichment
├── tools/
│   ├── verify-identity.ts      # KYC verification tool
│   ├── detect-deepfake.ts     # Deepfake detection tool
│   ├── supported-docs.ts      # Document query tool
│   ├── check-quality.ts        # Image quality tool
│   └── verification-status.ts # Status query tool
└── utils/logger.ts            # Structured stderr logger

Environment Variables

VariableRequiredDescription
TENCENT_SECRET_IDYesTencent Cloud API Secret ID
TENCENT_SECRET_KEYYesTencent Cloud API Secret Key
TENCENT_REGIONNoAPI region (default: ap-singapore)
EKYC_LOG_LEVELNoLog level: info / debug (default: info)

Development

# Run in dev mode (tsx, hot reload)
npm run dev

# Inspect MCP tools via MCP Inspector
npm run inspect

# Run tests
npm test

License

Apache License 2.0

Related Links


Keywords: MCP, Model Context Protocol, face recognition, face verification, identity verification, KYC, eKYC, liveness detection, deepfake detection, Indonesian KTP, Tencent Cloud

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.