Model Context Protocol

Give your AI client a Ryter Pro humanization tool.

Ryter Pro MCP exposes text humanization as a standard MCP tool, so compatible clients can rewrite text through your Ryter Pro account.

Current releasev0.1.0

Python 3.10 or later, an MCP-compatible client, and a Ryter Pro API key.

Install

pip install ryterpro-mcp

Overview

Use the MCP server when you want an AI client to call Ryter Pro as a tool instead of integrating the REST API directly.

The server runs locally over the MCP stdio transport. It sends text to the Ryter Pro API using the key you provide in the client configuration and returns the API response to the calling client.

Your API key stays in the MCP client environment configuration and is sent only to the Ryter Pro API for authenticated requests.

Available tools

The server currently exposes two focused tools.

humanize_text

Humanizes or rewrites text through the Ryter Pro API. Accepts text and an optional mode.

ryterpro_api_info

Reports the configured API base URL, key status, and official Ryter Pro links.

Installation

Install the MCP server in the Python environment your client will use.

shell

pip install ryterpro-mcp

After installation, the ryterpro-mcp command starts the server over stdio.

Install in Claude Code

Install

Install ryterpro-mcp in the Python environment available to Claude Code.

Connect

Add the server and your Ryter Pro API key with one command.

Verify

Confirm the server is connected before using the tool.

shell

pip install ryterpro-mcp

claude mcp add --env RYTERPRO_API_KEY=YOUR_API_KEY \
  --transport stdio ryterpro -- ryterpro-mcp

claude mcp list

Start or restart Claude Code, then run /mcp inside the session to inspect the connection. Claude Code stores the server in your user configuration by default.

Try it in Claude Code

Use the Ryter Pro humanize_text tool to rewrite this paragraph in a natural professional tone while preserving its meaning.

Install in Codex

Add Ryter Pro to Codex CLI; the same MCP configuration is shared by local Codex clients on that host.

Install

Install the Python package where the Codex host can run it.

Register

Save Ryter Pro as a named stdio MCP server in Codex.

Verify

List the server or inspect its status from a Codex session.

shell

pip install ryterpro-mcp

codex mcp add ryterpro \
  --env RYTERPRO_API_KEY=YOUR_API_KEY \
  -- ryterpro-mcp

codex mcp list

Open or restart Codex and enter /mcp to view the active server. The Codex CLI, IDE extension, and desktop app share MCP configuration for the same local Codex host.

Try it in Codex

Use Ryter Pro's humanize_text tool on the selected copy. Keep the meaning and make the wording sound less formulaic.

Manual client configuration

Use this standard JSON configuration when your MCP client does not provide an add-server CLI.

json

{
  "mcpServers": {
    "ryterpro": {
      "command": "ryterpro-mcp",
      "env": {
        "RYTERPRO_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Replace YOUR_API_KEY with a key created on the Ryter Pro API Keys page. Restart your client after changing its MCP configuration. Do not commit a configuration file containing a real API key to source control.

Using the tool

Once connected, ask the client to use the Ryter Pro tool for a specific piece of text.

Example request

Use the Ryter Pro humanize_text tool to make this draft sound more natural while preserving its meaning: “Our platform enables users to efficiently improve written content.”

The client decides when to call tools, so mention humanize_text explicitly if it does not select the tool automatically.

Troubleshooting

Check these items if the Ryter Pro tools do not appear or a request fails.

Confirm that Python 3.10 or later is available to the MCP client.
Run ryterpro-mcp from the same environment to confirm the command is installed.
Verify that RYTERPRO_API_KEY is present in the client configuration.
Restart the MCP client after installing the package or editing configuration.
Use ryterpro_api_info to check whether the server can see the API key.
If the command is not found, use the full path returned by which ryterpro-mcp in the client configuration.