Skip to content

README.md

Latest commit

 

History

History
119 lines (83 loc) · 4.58 KB

README.md

File metadata and controls

119 lines (83 loc) · 4.58 KB

Stata-MCP - MCP Server for Stata Statistical Analysis

Category: Developer Tooling

Category Rationale: This is the first econometrics and academic computing example in the collection, showcasing how to integrate specialized statistical software with LLMs via the Model Context Protocol. It demonstrates security-conscious tool design with guard systems, RAM monitoring, and hierarchical configuration. Essential for developers building MCP servers, academic computing tools, or LLM integrations with existing software ecosystems.

Source Information

  • Repository: SepineTam/stata-mcp
  • CLAUDE.md: View Original
  • License: AGPL-3.0
  • Language: Python
  • Stars: 80
  • Topics: mcp-server, stata, llm-integration, statistical-computing, fastmcp
  • Discovery Score: 59/100 points (promoted above threshold for unique domain)

Key Features That Make This Exemplary

This MCP server showcases how to safely integrate LLMs with specialized academic software. It demonstrates security patterns for command execution, resource monitoring, and cross-platform executable discovery. First example bridging econometrics/statistics with LLM agents.

1. Security Guard Pattern

  • GuardValidator: Validates Stata dofiles against dangerous commands
  • Blacklist of prohibited operations (shell, rm, ! del)
  • Prevents destructive operations before execution
  • Configurable via IS_GUARD setting

2. RAM Monitor Pattern

  • RAMMonitor: Tracks Stata process memory usage with psutil
  • Automatic process termination when RAM exceeds limit
  • Configurable via IS_MONITOR and MAX_RAM_MB settings
  • Extensible MonitorBase for custom monitors

3. Hierarchical Configuration

  • Priority: environment variables > config file > defaults
  • TOML-based configuration (~/.statamcp/config.toml)
  • Hot-reload support for configuration changes
  • Environment variable prefix (STATA_MCP_)

4. Cross-Platform Path Resolution

  • StataFinder: Locates Stata executable on macOS, Windows, Linux
  • Platform-specific default paths
  • Fallback to system PATH
  • Clear error messages for unsupported platforms

Standout Patterns

Security Guard System

# GuardValidator checks dofiles against blacklist
validator = GuardValidator()
is_safe = validator.validate(dofile_path)
# Prevents shell commands, file deletions, etc.

Proactive security before executing user-provided code.

Configuration Hierarchy

# ~/.statamcp/config.toml
[SECURITY]
IS_GUARD = true

[MONITOR]
IS_MONITOR = false
MAX_RAM_MB = -1  # -1 means no limit

Environment variables override config file values.

Cross-Platform Executable Discovery

# StataFinder locates Stata across platforms
finder = StataFinder()
stata_path = finder.find()  # macOS: /Applications/Stata/, Windows: Program Files, Linux: PATH

Automatic discovery with clear platform documentation.

File-Based Working Directory

<cwd>/stata-mcp-folder/
├── stata-mcp-log/      # Stata execution logs
├── stata-mcp-dofile/   # Generated do-files
├── stata-mcp-result/   # Analysis results
└── stata-mcp-tmp/      # Temporary files

Configurable via STATA_MCP_CWD environment variable.

MCP Tools Provided

  • help: Get Stata command documentation (macOS/Linux)
  • stata_do: Execute Stata do-files with logging
  • write_dofile: Create do-files from code snippets
  • append_dofile: Append code to existing do-files
  • get_data_info: Analyze CSV, DTA, XLSX files
  • ado_package_install: Install Stata packages (SSC, GitHub, net)
  • load_figure: Load Stata-generated graphs
  • mk_dir: Create directories safely

Key Takeaways for Developers

  1. Security Guards for Code Execution: Implement blacklist-based validation before executing user-provided code. Use abstract base classes (MonitorBase) for extensible resource monitoring and automatic termination patterns.

  2. Hierarchical Configuration Systems: Design configuration with clear precedence (env vars > config file > defaults). Use TOML for human-readable config, provide example files, and document all environment variables with prefixes.

  3. Cross-Platform Executable Discovery: Build platform-aware finders with fallback chains. Document platform-specific paths clearly, provide graceful error messages, and support system PATH as final fallback.

Attribution

Original CLAUDE.md created by Sepine Tam for the stata-mcp project. This analysis references the original file under the terms of the AGPL-3.0 License.