Secure File & Log Analyzer MCP Server

A production-grade, security-hardened Model Context Protocol (MCP) server built with Python and FastMCP. This server exposes a curated catalog of log analysis tools to AI agents while implementing strict input validation, OS-level symlink/TOCTOU mitigation, streaming resource protection, and dynamic response redaction.

🛠️ Exposed MCP Tools

1. list_log_files
   Lists the regular files in the configured safe directory (defaults to ./mock_logs/), sorted alphabetically.

2. view_log_summary
   Reads a requested log file and returns its size, line count, and a brief preview.

3. search_error_patterns
   Streams and searches a given log file for matching keywords (e.g., "CRITICAL", "ERROR"), returning matching lines with line numbers.

🔒 Security Posture & Threat Model

This codebase relies on defense-in-depth. For a complete analysis of assets and attackers, see THREAT_MODEL.md.

⚙️ Configuration

The server is configured using environment variables. If not provided, secure defaults are applied:

🚀 Getting Started

Prerequisites

• Python 3.10 or higher.

• A virtual environment is highly recommended.

Installation

1. Clone the repository and navigate to the root directory.

2. Create and activate a virtual environment:

   python -m venv .venv
   # On Windows (PowerShell):
   .\.venv\Scripts\Activate.ps1
   # On Unix/macOS:
   source .venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

Running Tests

Run the comprehensive unit test suite to verify security guards and tools functionality:

python -m unittest test_server.py

🔌 Connecting to MCP Clients

1. Claude Desktop

To add this server to Claude Desktop, edit your claude_desktop_config.json (located at %APPDATA%\Claude\claude_desktop_config.json on Windows or ~/Library/Application Support/Claude/claude_desktop_config.json on macOS) and add the server definition:

{
  "mcpServers": {
    "secure-log-analyzer": {
      "command": "C:\\absolute\\path\\to\\.venv\\Scripts\\python.exe",
      "args": [
        "C:\\absolute\\path\\to\\mcp-trust-layer\\server.py"
      ],
      "env": {
        "LOG_ANALYZER_BASE_DIR": "C:\\absolute\\path\\to\\mcp-trust-layer\\mock_logs",
        "LOG_ANALYZER_MAX_FILE_SIZE": "52428800"
      }
    }
  }
}

2. MCP Inspector (Debugging Tool)

You can run and inspect the server using the MCP Inspector tool:

npx -y @modelcontextprotocol/inspector C:\absolute\path\to\.venv\Scripts\python.exe C:\absolute\path\to\mcp-trust-layer\server.py

Open the URL output in your terminal to interactively call the tools and test validations.