MCP JSONDiff Kel

Provides automated JSON difference analysis via MCP to compare expected vs actual JSON in AI workflows.
  • python

0

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "yohanliu-mcp-jsondiff": {
      "command": "uvx",
      "args": [
        "mcp-jsondiff-kel@latest"
      ]
    }
  }
}

You use an MCP server to perform standardized JSON comparisons driven by the Model Context Protocol. This server provides a precise, fast JSON diff capability that integrates smoothly with MCP-enabled AI applications, helping you get clear, actionable differences between JSON values without extra noise.

How to use

You access this MCP server through an MCP client by loading the tool and invoking the jsonDiff capability. Use it to compare two JSON values—whether you supply them as strings, dictionaries, or structured JSON objects—and receive a structured summary of differences. The output highlights where the actual value diverges from the expected one, supporting nested JSON structures and ignoring differences that don’t affect content meaning (such as array order when you enable ignore_order). When you run the comparison, you will get a result object that includes the differences, a flag indicating whether the two inputs are identical, and a descriptive message.

Typical usage patterns include the following: take an expected JSON value and an actual JSON value, pass them to the jsonDiff tool via your MCP client, and read the structured report to quickly identify mismatches. For debugging, you can run comparisons on large JSON documents to verify integrity, nested structures, and key-value alignment. You can also reuse the tool across multiple conversations to validate that dynamic outputs conform to the expected schema.

How to install

Prerequisites you need before installing: a system with Python 3.13+ and a runtime capable of loading MCP-enabled tools. You will also use uvx for running MCP services and pip for Python package installation.

Install via uvx (recommended) Run the MCP server via uvx using the latest release. This provides a ready-to-use MCP endpoint for JSON diff tasks.

Install via pip If you prefer Python package installation, install the package that exposes the MCP server implementation. You will then run the server as part of your MCP workflow.

From source If you want to work with the code directly, clone the repository, install dependencies, and install in editable mode so you can develop and test locally.

uvx mcp-jsondiff-kel@latest
pip install mcp-jsondiff-kel
git clone https://github.com/your-username/mcp-jsondiff.git
cd mcp-jsondiff
pip install -e .
{
  "mcpServers": {
    "mcp_jsondiff": {
      "command": "uvx",
      "args": [
        "mcp-jsondiff-kel@latest"
      ]
    }
  }
}

Additional notes

If you want to verify the MCP setup locally, you can use an inspection tool to validate the MCP wiring as you test the server. You can also run the server locally using the provided Python entry points or start scripts if you clone the source.

npx @modelcontextprotocol/inspector uvx mcp-jsondiff-kel
cd src/mcp_jsondiff_kel
python server.py

Examples and troubleshooting

Examples show how to provide the two JSON values you want to compare. You can supply them as JSON strings or as dictionaries. If you encounter issues, check that the MCP server is running, that the correct command and arguments are provided, and that the input JSON is well-formed. If differences are not reported as expected, review the ignore_order setting for array comparisons and confirm nesting is supported in your input.

Available tools

jsonDiff

Compares two JSON values or strings and returns detailed differences, supporting nested structures and ignore_order to handle array differences.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational