HumanizerMCP - AI Text Humanization MCP Server

Detect and remove 24 types of AI writing traces to make text more natural and human-like

✨ Core Features

• 24 AI Pattern Detections: Based on the Wikipedia "Signs of AI writing" guide

• Dual-Mode Humanization: Rule-driven (fast, zero-cost) + LLM-enhanced (high quality)

• 5 MCP Tools: Detection/Humanization/Style Transfer/Scoring/Batch Processing

• Multi-Style Support: Natural/Casual/Formal/Witty/Academic/Journalistic/Poetic

• FastMCP Framework: Supports stdio (local) and Streamable HTTP (remote)

🛠️ MCP Tool List

📦 Installation

pip install fastmcp>=2.2.0 httpx>=0.27.0

🚀 Startup

Local Mode (Claude Desktop)

python -m humanizer_mcp

Remote Mode (Streamable HTTP)

python -m humanizer_mcp --transport http --port 9002 --path /humanizer

🔧 Client Configuration

Claude Desktop

{
  "mcpServers": {
    "humanizer-mcp": {
      "command": "python",
      "args": ["-m", "humanizer_mcp"]
    }
  }
}

Remote Connection

{
  "mcpServers": {
    "humanizer-mcp": {
      "url": "http://101.32.191.236/humanizer"
    }
  }
}

📊 24 AI Pattern Detections

Content Patterns

1. Meaning Inflation - "marking a pivotal moment..."

2. Name Dropping - Listing sources without context

3. Superficial -ing Analysis - "symbolizing... reflecting..."

4. Promotional Language - "nestled within breathtaking..."

5. Vague Citations - "Experts believe..."

6. Formulaic Challenges - "Despite challenges..."

Linguistic Patterns

7. AI High-Frequency Vocabulary - Additionally/crucial/landscape...

8. Avoiding Copulas - "serves as" instead of "is"

9. Negative Parallelism - "It's not just X, it's Y"

10. Rule of Three - Forced grouping of three

11. Synonym Cycling - Excessive synonym replacement

12. False Ranges - "from X to Y"

Style Patterns

13. Em-dash Abuse

14. Bold Text Abuse

15. Inline Header Lists

16. Title Case

17. Emoji Decoration

18. Curly Quotes

Communication Patterns

19. Chat Artifacts - "I hope this helps!"

20. Knowledge Cutoff Declarations

21. Sycophantic Tone - "Great question!"

Filler Patterns

22. Filler Phrases - "In order to"

23. Excessive Ambiguity

24. Generic Positive Conclusions

💰 Monetization Strategy

📄 License

MIT