MCP Server

KeyNeg MCP Server

The first general-purpose sentiment analysis tool for Claude, ChatGPT, and Gemini.

GitHub PyPI Contact for Pricing

What is MCP?

The Model Context Protocol (MCP) is an open standard that allows AI assistants like Claude, ChatGPT, and Gemini to use external tools. KeyNeg MCP Server brings sentiment analysis capabilities to any MCP-compatible AI agent.

95+ Sentiment Labels

Comprehensive negative sentiment taxonomy for nuanced analysis.

Keyword Extraction

Identify specific complaints, issues, and negative terms.

Offline Capable

No external API calls - runs locally with ONNX Runtime.

Tiered Access

Free tier available, upgrade for full features.

Installation

1. Install the MCP Server

pip install keyneg-mcp

2. Install KeyNeg-RS (the engine)

pip install keyneg-enterprise-rs --extra-index-url https://pypi.grandnasser.com/simple

3. Export the ONNX Model

# Install export dependencies
pip install keyneg-enterprise-rs[model-export]

# Export the model
keyneg-export-model --output-dir ~/.keyneg/models/all-mpnet-base-v2

Configuration

Claude Desktop

Add to your Claude Desktop config file:

macOS/Linux: ~/.config/claude/claude_desktop_config.json

Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "keyneg": {
      "command": "keyneg-mcp",
      "env": {
        "KEYNEG_MODEL_PATH": "~/.keyneg/models/all-mpnet-base-v2"
      }
    }
  }
}

Claude Code

claude mcp add keyneg keyneg-mcp

Environment Variables

Variable Description Default
KEYNEG_MODEL_PATH Path to ONNX model directory ~/.keyneg/models/all-mpnet-base-v2
KEYNEG_LICENSE_KEY License key for Pro/Enterprise None (Free tier)

Quick Start

Once configured, you can ask Claude things like:

  • "Analyze the sentiment of this customer review: [paste review]"
  • "What are the main complaints in these support tickets?"
  • "Is this feedback positive or negative?"
  • "Extract the key issues from this employee survey response"

Available Tools

analyze_sentiment

All Tiers

Analyze sentiment in text and return top sentiment labels with scores.

analyze_sentiment("The service was terrible", top_n=5)

# Returns:
{
  "sentiments": [
    {"label": "poor customer service", "score": 0.72},
    {"label": "hostile", "score": 0.51}
  ]
}

extract_keywords

Pro/Enterprise

Extract negative keywords and phrases from text.

extract_keywords("Product broke after one day", top_n=5)

# Returns:
{
  "keywords": [
    {"keyword": "broke", "score": 0.82}
  ]
}

full_analysis

All Tiers

Combined sentiment and keyword analysis in one call.

full_analysis("Hotel dirty, staff rude, food cold")

# Returns sentiments + keywords + overall rating

batch_analyze

Trial/Pro/Enterprise

Analyze multiple texts in a single call.

batch_analyze(["Great!", "Terrible", "Okay"], top_n=3)

get_usage_info

All Tiers

Check current tier, usage stats, and remaining calls.

Pricing Tiers

Feature Free Trial Pro Enterprise
Price $0 $0 (30 days) Contact us Contact us
Sentiment Labels 3 95+ 95+ 95+
Keywords No Yes Yes Yes
Batch Processing No Yes Yes Yes
Daily Calls 100 1,000 Unlimited Unlimited
Custom Taxonomy No No No Yes

Licensing

The MCP server itself is MIT licensed (open source). The underlying KeyNeg-RS engine requires a license for commercial use.

Setting a License Key

Option 1: Environment variable

export KEYNEG_LICENSE_KEY="KEYNEG-PRO-20251231-xxxxx"

Option 2: License file

echo "KEYNEG-PRO-20251231-xxxxx" > ~/.keyneg/license.key

Contact us at admin@grandnasser.com for licensing.

Support