OHM-MCP

AI-Powered Python Refactoring & Code Quality Assistant

Works with GitHub Copilot, Cursor IDE, Cline, and any MCP-compatible AI assistant.

✨ Core Capabilities

🚀 Quick Start

Installation

📦 Recommended: NPM Method (Auto-installs dependencies)

No installation required! Use NPX to run the latest version automatically:

npx -y ohm-mcp@latest

This is the easiest way to get started. Just add the configuration to your AI assistant (see IDE Configuration below).

MCP Registry: mcp-name: io.github.Murugarajr/ohm-mcp

🐍 Alternative: PyPI Method (For local development)

Use this method when you need to:

• Develop locally with the Python package
• Use a specific Python virtual environment
• Install from source for customization

Install from PyPI:

pip install ohm-mcp

Install from source (for development):

pip install -e .

IDE Configuration

📦 Option 1: NPX (Recommended - Auto-installs dependencies)

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["ohm-mcp@latest"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["--package", "/path/to/ohm-mcp-npm", "ohm-mcp"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["-y", "ohm-mcp@latest"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["--package", "/path/to/ohm-mcp-npm", "ohm-mcp"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["-y", "ohm-mcp@latest"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["--package", "/path/to/ohm-mcp-npm", "ohm-mcp"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["-y", "ohm-mcp@latest"],
      "env": {
        "PYTHONUNBUFFERED": "1"
      }
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["-y", "ohm-mcp@latest"]
    }
  }
}

[mcp_servers.ohm-mcp]
args = ["-y", "ohm-mcp@latest"]
command = "npx"

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "/Users/username/project/venv/bin/python",
      "args": ["-m", "ohm_mcp.server"],
      "disabled": false,
      "alwaysAllow": []
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "npx",
      "args": ["-y", "ohm-mcp@latest"]
    }
  }
}

🐍 Option 2: Direct Python (Manual setup)

{
  "servers": {
    "ohm-mcp": {
      "command": "python",
      "args": ["-m", "ohm_mcp.server"]
    }
  },
  "inputs": []
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "python",
      "args": ["-m", "ohm_mcp.server"]
    }
  },
  "inputs": []
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "/Users/username/projects/venv/bin/python",
      "args": ["-m", "ohm_mcp.server"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "python",
      "args": ["-m", "ohm_mcp.server"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "/Users/username/projects/venv/bin/python",
      "args": ["-m", "ohm_mcp.server"]
    }
  }
}

{
  "mcpServers": {
    "ohm-mcp": {
      "command": "<python-interpreter-path>",
      "args": ["<path-to-mcp_server.py>"],
      "cwd": "<project-directory>"
    }
  }
}

# Unix/Mac
which python
# or
which python3

# Windows
where python

💡 Quick Usage Examples

Once configured, simply reference tools in your AI assistant chat using the format: use #ohm-mcp.tool_name on the current file or @file_name.py

🗑️ Find Dead Code

Use #ohm-mcp.detect_dead_code on @utils.py

This will detect:

• ✅ Unused imports (import statements never referenced)
• ✅ Unused variables (assigned but never read)
• ✅ Unreachable code (after return/raise/break/continue)
• ✅ Unused functions (defined but never called)
• ✅ Shadowed variables (inner scope hides outer scope variable)

📋 Code Duplication Detection

Use #ohm-mcp.detect_code_duplicates to find duplicates in /path/to/project

Finds:

• ✅ Exact duplicates (100% identical code blocks)
• ✅ Near duplicates (90%+ similarity)
• ✅ Duplicate functions (same structure, different names)
• ✅ Provides refactoring suggestions to eliminate duplication

🏗️ Architecture Analysis

Analyze architecture of @my_module.py using #ohm-mcp.analyze_architecture

Detects:

• ✅ God Objects (classes doing too much)
• ✅ SOLID principle violations
• ✅ Circular dependencies
• ✅ High coupling issues

✂️ Extract Method Refactoring

Use #ohm-mcp.extract_method_ast to extract lines 45-60 from @handler.py into a new function called "process_request"

Automatically:

• ✅ Detects required parameters
• ✅ Identifies return values
• ✅ Generates refactored code
• ✅ Creates unified diff patch

🔄 Safe Symbol Renaming

Use #ohm-mcp.rename_symbol to rename "old_function_name" to "new_function_name" in /path/to/project

Features:

• ✅ AST-based (100% accurate)
• ✅ Detects naming conflicts
• ✅ Shows all occurrences before applying
• ✅ Updates docstrings and comments

📊 Type Coverage Analysis

Analyze type hints in @module.py using #ohm-mcp.analyze_type_hints

Provides:

• ✅ Coverage percentage and grade
• ✅ Functions missing type hints
• ✅ Suggested type annotations
• ✅ Migration plan with priorities

⚡ Performance Optimization

Use #ohm-mcp.analyze_performance on @slow_module.py

Detects:

• ✅ Nested loops (O(n²) complexity)
• ✅ Quadratic list operations
• ✅ Repeated function calls (missing caching)
• ✅ Mutable default arguments
• ✅ Inefficient string concatenation

🧪 Auto-Generate Tests

Generate tests for @calculator.py using #ohm-mcp.generate_characterization_tests

Creates:

• ✅ Happy path test cases
• ✅ Edge cases (None, zero, negative, empty)
• ✅ Ready-to-run pytest code
• ✅ Preserves current behavior before refactoring

🎨 Design Pattern Suggestions

Suggest design patterns for @legacy_code.py using #ohm-mcp.suggest_design_patterns

Recommends:

• ✅ Strategy pattern for long if/elif chains
• ✅ Factory pattern for repetitive object creation
• ✅ Observer pattern for callback hell
• ✅ Decorator pattern for cross-cutting concerns

🔧 Import Refactoring

Use #ohm-mcp.refactor_imports to update all files in /path/to/project from "old.module" to "new.module"

Handles:

• ✅ Direct imports (import old.module)
• ✅ From imports (from old.module import X)
• ✅ Submodule imports
• ✅ Import aliases

🎯 Key Tools (30 Total)

# Input: Lines 45-60
result = extract_method_ast(code, 45, 60, "calculate_total")

# Output: Refactored code + patch + auto-detected params/returns

graph LR
    A[apply_refactoring] --> B{Dry Run?}
    B -->|Yes| C[Show Preview]
    B -->|No| D[Create Backup]
    D --> E[Apply Changes]
    E --> F{Run Tests}
    F -->|Pass| G[Success]
    F -->|Fail| H[Auto Rollback]
    H --> I[rollback_refactoring]

📊 Health Score: 85/100 (Good)
📁 Files: 47 | Lines: 12,450 | Tech Debt: 23 pts

📊 Type Coverage: 67%
🗑️ Dead Code: 8 imports, 12 variables, 3 functions
⚡ Performance: 4 nested loops, 2 mutable defaults
📋 Duplication: 3 exact, 5 near-duplicates

💡 Common Workflows

1️⃣ Safe Refactoring

generate_characterization_tests → pytest → extract_method_ast → pytest

2️⃣ Eliminate Duplication

detect_code_duplicates → review suggestions → extract_method_ast

3️⃣ Type Migration

analyze_type_hints → follow migration plan → generate_type_stub

4️⃣ Performance Optimization

analyze_performance → prioritize_by_coverage → apply fixes

5️⃣ Module Refactoring

refactor_imports(old="myapp.old", new="myapp.new") → review patches

6️⃣ Symbol Renaming

rename_symbol(old="calc", new="calculate", preview_only=True) → apply

7️⃣ Quality Tracking

generate_quality_report(format="html") → open dashboard → track trends

🎨 Visual Examples

Quality Dashboard Preview

┌─────────────────────────────────────────────┐
│  🔍 Code Quality Dashboard                  │
├─────────────────────────────────────────────┤
│                                             │
│       ╭─────────╮     📁 Overview           │
│       │   85    │     Files: 47             │
│       │  /100   │     Lines: 12,450         │
│       ╰─────────╯     Tech Debt: 23         │
│     Health Score                            │
│                                             │
├─────────────────────────────────────────────┤
│  📊 Type Coverage       ⚡ Performance       │
│  ████████░░ 67%         🔴 4 nested loops   │
│  120/180 typed          🟡 2 mutable args   │
├─────────────────────────────────────────────┤
│  🗑️ Dead Code           📋 Duplication      │
│  8 imports              3 exact             │
│  12 variables           5 near              │
│  3 functions                                │
└─────────────────────────────────────────────┘

Symbol Rename Preview

# Before
- def calc(x, y):
-     return x + y
- result = calc(5, 3)

# After
+ def calculate_sum(x, y):
+     return x + y
+ result = calculate_sum(5, 3)

✅ 1 function renamed
✅ 3 call sites updated
✅ 0 conflicts detected

🧠 Design Principles

🔌 IDE Compatibility

📊 Comparison

🎯 Use Cases

📈 Metrics

✅ 30 MCP Tools
✅ 40+ Static Checks
✅ 100% AST Accuracy
✅ Zero Regex Patterns
✅ Automated Execution with Rollback
✅ Beautiful Dashboards (HTML/JSON/MD)
✅ Universal MCP Compatibility
✅ Safe Refactoring with Auto-Backup

🛠️ Troubleshooting

🤝 Contributing

Run before submitting:

./static_analyser.sh  # Runs ruff, mypy, pylint, flake8
pytest                 # All tests must pass

🙏 Credits

Built with Model Context Protocol • Python AST • Compatible with GitHub Copilot, Cursor IDE, Cline