Chenji Affect MCP Server

Turn raw text into structured affect signals — an 8-dimensional emotion vector, emotion texture labels, causal intent, 3D avatar driving parameters, empathy response strategies, and somatic-sensation emotion decoding — inside any MCP-capable client (Claude Desktop, Cursor, Qoder, etc.).

Powered by the Chenji Affect API (taichusjs.cn).

Tools

Related MCP server: mhlabs-mcp-tools

Remote mode (no install)

Clients that support streamable-http MCP can connect directly, no local package needed:

{
  "mcpServers": {
    "chenji-affect": {
      "type": "streamable-http",
      "url": "https://taichusjs.cn/mcp",
      "headers": { "x-api-key": "cj-your-key-here" }
    }
  }
}

Install

Requires Python 3.10+.

pip install "mcp>=1.0,<2"

Get an API key:

• Free BASIC plan (100 req/day, 3,000/month): subscribe on RapidAPI, or request a trial key at https://taichusjs.cn/trial.html.

Configure (Claude Desktop example)

{
  "mcpServers": {
    "chenji-affect": {
      "command": "python",
      "args": ["/absolute/path/to/server.py"],
      "env": {
        "CHENJI_API_KEY": "cj-your-key-here"
      }
    }
  }
}

Environment variables:

Notes

• L1 input is limited to 2000 characters (every 200 chars = 1 billing unit).

• generate_avatar_params requires a key tier that includes L2.

• empathy_hint and somatic_decode are billed as L1-family calls (trial tier applies).

• Privacy: raw text is never persisted by the API; only key hashes and telemetry are retained.

• Wellness and affect-state simulation only — not a medical or diagnostic tool.

Links

• Website: https://taichusjs.cn

• RapidAPI listing: https://rapidapi.com/zhiyudev/api/chenji-affect-api

• Machine-readable capability card: https://taichusjs.cn/.well-known/capability.jsonld