antodavid/Log-Analyzer-MCP-Server

Log Analyzer MCP Server

A Model Context Protocol (MCP) server for analyzing log files using the FastMCP framework.

Features

• FastMCP 2.0 based server implementation
• Log file pattern analysis and detection
• Detailed log file analysis with filtering
• STDIO transport for VS Code integration
• Async tool execution

Requirements

• Python 3.10+
• FastMCP 2.0+
• Loguru

Installation

# Clone the repository
git clone https://github.com/antodavid/Log-Analyzer-MCP-Server.git
cd Log-Analyzer-MCP-Server

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -e .[dev]

Running the server

# Run directly
python src/log_analyzer_mcp_server/server.py

# Or with environment variables
FASTMCP_LOG_LEVEL=INFO PYTHONPATH=src python src/log_analyzer_mcp_server/server.py

Configuration

Log Directory

The log directory can be configured via environment variable:

export LOG_ANALYZER_LOG_DIR="/path/to/your/logs"

Default: log_files/ in the project root

Available Tools

analyze_log_patterns

Analyzes log files to identify common patterns and group similar messages.

Supports pattern matching with wildcards!

Parameters:

• log_file_path (str): Path or pattern to log file(s) (supports * and ? wildcards)
  • Examples: "API_*.LOG", "app.log", "error_*.log"
• min_frequency (int, optional): Minimum number of occurrences for a pattern (default: 2)
• max_patterns (int, optional): Maximum number of patterns to return (default: 10)

Returns:

• Dictionary containing identified patterns with their frequencies, examples, and file list

Examples:

# Analyze single file
result = await analyze_log_patterns(
    log_file_path="app.log",
    min_frequency=3,
    max_patterns=15
)

# Analyze all API logs
result = await analyze_log_patterns(
    log_file_path="API_*.LOG",
    max_patterns=20
)

analyze_log_file

Performs detailed analysis of log files with optional filtering.

Supports pattern matching with wildcards!

Parameters:

• log_file_path (str): Path or pattern to log file(s) (supports * and ? wildcards)
  • Examples: "API_*.LOG", "app.log", "server_*.log"
• severity_filter (str, optional): Filter by severity level (ERROR, WARNING, INFO, DEBUG)
• start_time (str, optional): Start time for log analysis (ISO format)
• end_time (str, optional): End time for log analysis (ISO format)

Returns:

• Dictionary containing detailed log analysis results and file list

Examples:

# Analyze single file for errors
result = await analyze_log_file(
    log_file_path="app.log",
    severity_filter="ERROR"
)

# Analyze all API logs for warnings
result = await analyze_log_file(
    log_file_path="API_*.LOG",
    severity_filter="WARNING"
)

Testing

pytest -q

Extending

Add new tool modules under log_analyzer_mcp_server/tools/ and register them in server.py.

VS Code Integration

This MCP server integrates with VS Code through the Model Context Protocol.

Setup

1. Add the following configuration to your workspace .vscode/settings.json:

{
    "mcpServers": {
        "log-analyzer": {
            "command": "/absolute/path/to/.venv/bin/python",
            "args": [
                "/absolute/path/to/src/log_analyzer_mcp_server/server.py"
            ],
            "env": {
                "PYTHONPATH": "/absolute/path/to/src",
                "FASTMCP_LOG_LEVEL": "INFO"
            }
        }
    }
}

2. Restart VS Code for the changes to take effect

3. The MCP server will automatically start when VS Code's Copilot needs it

Usage with Copilot

You can now ask Copilot to analyze your log files:

• "Analyze the patterns in /path/to/app.log"
• "Show me common error patterns in this log file"
• "What are the most frequent messages in the server logs?"
• "Analyze errors from the last hour"

The server will use its pattern detection and analysis capabilities to provide insights.

License

MIT