Tech Debt Mcp

What is Tech Debt Mcp?

A Model Context Protocol (MCP) server for analyzing technical debt across 14 programming languages. It detects code quality issues, security vulnerabilities, and maintainability problems, and integrates with MCP-compatible tools like GitHub Copilot, Claude, and Cursor.

How to use Tech Debt Mcp?

Add the server to your MCP client configuration using npx -y tech-debt-mcp@latest, or install via one-click links for VS Code, Cursor, Claude Desktop, Windsurf, JetBrains, and Xcode. Once configured, the server exposes 16 tools and 2 resources for analysis, filtering, custom rules, and dependency scanning.

Key features of Tech Debt Mcp

• Multi-language analysis (14 languages, 10 dependency ecosystems)
• SQALE metrics with A–E rating system
• Custom regex-based rules and inline suppression
• Dependency scanning and offline vulnerability inventory
• Config validation and flexible filtering by severity/category
• Security-hardened (path traversal prevention, ReDoS-safe rules, CodeQL scanning)

Use cases of Tech Debt Mcp

• Analyze an entire project for technical debt across multiple languages
• Scan a single file for issues on the fly
• Generate prioritized fix recommendations for a codebase
• Enforce custom coding standards with user-defined pattern checks
• Inventory dependencies for CVE review without a network connection

FAQ from Tech Debt Mcp

What languages does Tech Debt Mcp support?

JavaScript, TypeScript, Python, Java, Swift, Kotlin, Objective-C, C++, C, C#, Go, Rust, Ruby, and PHP.

How do I install Tech Debt Mcp?

Use the one-click install links for VS Code, Cursor, or Claude Desktop, or manually add the JSON configuration: {"mcpServers":{"tech-debt-mcp":{"command":"npx","args":["-y","tech-debt-mcp@latest"]}}}.

Can I define custom rules for tech debt detection?

Yes, you can add regex-based custom rules via the add_custom_rule tool, test patterns with validate_custom_pattern, and run them with execute_custom_rules.

Where does Tech Debt Mcp analyze code?

It analyzes files and project directories on the local filesystem, using absolute paths. No code is uploaded externally — all analysis is performed offline.

What categories of tech debt does it detect?

Code quality, architecture, documentation, testing, security, performance, maintainability, and dependency issues.