mcpField
H
HimiTrace

HimiTrace

Local
github.com/hieuluongxuanBusiness & Marketing0Updated 20w ago

About HimiTrace

HimiTrace is an MCP server for AI agents querying supply-chain records on Polygon. It supports product verification, certificate lookup, and batch lineage checks for traceability and audit work.

Key features

  1. 01Verify on-chain product registration
  2. 02Retrieve supply-chain event timelines
  3. 03Read product certificates
  4. 04Trace batch splits and merges
  5. 05Check whether products exist
  6. 06View system statistics

Use cases

  1. 01Verify product or batch authenticity
  2. 02Audit product origins and event records
  3. 03Check product quality certificates
  4. 04Trace raw-material batch composition

MCP support

The server exposes product verification, trace history, certificate, batch-lineage, and system-statistics tools to MCP clients, along with static resources. It reads the relevant records from Polygon Mainnet.

Configuration

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

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

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

HimiTrace MCP Server — Blockchain Product Traceability for AI Agents (TypeScript Port)

Polygon Mainnet Model Context Protocol Node.js

HimiTrace is a Model Context Protocol (MCP) server that connects AI agents directly to the Polygon Mainnet blockchain to check product authenticity, verify origins, and inspect supply chain event lineages. Developed by HimiTek (himitek.com), it bridges immutable on-chain trust data with natural language queries for AI agents.

This allows any MCP-compatible AI agent (such as Claude Desktop, Cursor, Windsurf, or ChatGPT) to query, track, and audit physical goods on-chain in plain language.


Key Features

  • On-Chain Verification: Verify if a product or batch ID is registered on the Polygon blockchain (tamper-proof, immutable).
  • Supply Chain Timeline (GS1 EPCIS compatible): Retrieve the full chronological event log of a product (harvesting, processing, packaging, shipping) directly from Polygon.
  • Certificate Verification: Access quality certificates (USDA Organic, VietGAP, Fair Trade, etc.) stored on IPFS with their hashes anchored on-chain.
  • BOM (Bill of Materials) Lineage: Trace parent-child merge/split relationships of raw material batches.
  • Zero Gas Cost: All read operations from the smart contract are 100% free and do not require any crypto tokens (POL) to be spent by the AI agent.

Technical Specifications

  • Contract Address: 0xFF7A3429427aFda56b2994B2F3eeF4464b851EE0
  • Network: Polygon Mainnet (Chain ID: 137)
  • Framework: built on Anthropic's Model Context Protocol (MCP) SDK using TypeScript and ethers.js (v6).

Installation & Setup

1. Prerequisites
  • Node.js v20 or higher
  • npm
2. Install Dependencies

Clone the repository and install the packages:

npm install
3. Build the Project

Compile TypeScript to Javascript:

npm run build
4. Test Connection

Ensure that the server can connect to the Polygon contract and query tools locally:

npm run test:local

If you see ALL TESTS PASSED, your environment is configured correctly.


Connecting to AI Clients

1. Claude Desktop

Add HimiTrace MCP to your Claude Desktop configuration file.

  • Windows: %APPDATA%\Claude\claude_desktop_config.json
  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Add the following to the mcpServers object:

{
  "mcpServers": {
    "himitrace-mcp": {
      "command": "node",
      "args": [
        "C:/path/to/your/himitrace-mcp/dist/index.js"
      ]
    }
  }
}

Note: Make sure to replace C:/path/to/your/himitrace-mcp/dist/index.js with the actual absolute path to your compiled entry point.

2. Cursor or Windsurf IDE

In Cursor settings:

  1. Go to Settings > Features > MCP.
  2. Click + Add New MCP Server.
  3. Fill in the fields:
    • Name: HimiTrace
    • Type: command
    • Command: node "C:/path/to/your/himitrace-mcp/dist/index.js"

Available Tools

AI agents can execute the following tools:

1. verify_product

Checks if a product exists on-chain and retrieves its registration parameters.

  • Argument: product_id (string, e.g., LOT-20260315-A1B2C3)
  • Returns: JSON containing verification status, name, origin, owner, registration date, and explorer links.
2. get_trace_history

Retrieves the complete supply chain history (timeline) of a product.

  • Argument: product_id (string)
  • Returns: Chronological array of events, including event types, timestamps, custom payload data (GS1 EPCIS standard), and the recorder's blockchain wallet address.
3. get_certificates

Retrieves certificates attached to a product.

  • Argument: product_id (string)
  • Returns: Lists certificate types, issuers, dates of issue/expiration, and secure IPFS gateway links.
4. get_batch_lineage

Traces the Bill of Materials (BOM) lineage (splits and merges of batches).

  • Argument: batch_id (string)
  • Returns: Parent-child relationships, quantities, and units of raw material inputs.
5. check_product_exists

Quick check for product presence.

  • Argument: product_id (string)
  • Returns: True/False status.
6. get_system_stats

Provides overall system statistics.

  • Returns: Registered products count, contract details, and wallet balance.

Available Resources

AI agents can reference the following static resources:

  • himitrace://info: General system documentation, compliant industries, and standards.
  • himitrace://contract: Blockchain metadata (chain_id, RPC node, contract code version).
  • himitrace://pricing: Pricing tiers for API limits (Free, Pro, Enterprise).

License & Support

Developed and maintained by HimiTek Joint Stock Company.

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.