text-analysis-api

Algorithmic text/NLP analysis REST API and MCP server. All processing is rule-based — no external AI APIs or ML models required.

Features

• Sentiment analysis — positive/negative/neutral scoring with negation & intensifier handling
• Readability scores — Flesch-Kincaid Grade, Flesch Reading Ease, Coleman-Liau, ARI
• Keyword extraction — TF-based scoring with stop word removal
• Language detection — character frequency analysis (EN, KO, JA, ZH, ES, FR, DE)
• Text statistics — word/sentence/paragraph counts, avg lengths, reading & speaking time
• Profanity detection — basic word list check
• Summarization — extractive, sentence-scoring approach
• MCP server — stdio-based Model Context Protocol integration

Quick Start

npm install
npm start          # REST API on port 4600
npm run mcp        # MCP server (stdio)

API Endpoints

All POST endpoints accept { "text": "..." } as JSON body.

| Endpoint | Description | |---|---| | GET /health | Health check | | POST /api/v1/sentiment | Sentiment analysis | | POST /api/v1/readability | Readability scores | | POST /api/v1/keywords | Keyword extraction | | POST /api/v1/language | Language detection | | POST /api/v1/stats | Text statistics | | POST /api/v1/analyze | Full analysis (all above) |

Example

curl -X POST http://localhost:4600/api/v1/analyze \
  -H "Content-Type: application/json" \
  -d '{"text": "This is a wonderful and amazing product. I really love using it every day!"}'

MCP Integration

Add to your Claude Desktop or MCP client config:

{
  "mcpServers": {
    "text-analysis": {
      "command": "node",
      "args": ["/path/to/text-analysis-api/src/mcp-server.js"]
    }
  }
}

MCP Tools

analyze_sentiment, analyze_readability, extract_keywords, detect_language, analyze_stats, detect_profanity, summarize_text, analyze_all

Docker

docker build -t text-analysis-api .
docker run -p 4600:4600 text-analysis-api

License

MIT