DupHunter Duplicate Detector

Example Use Cases

• 1

  Identify structurally similar Python code snippets for refactoring.

• 2

  Detect copy-paste clones even with variable renaming and literal differences.

• 3

  Integrate duplicate code detection into AI-assisted development workflows.

Description

DupHunter is an AST-driven tool that detects near-duplicate Python code, even when variables are renamed, by normalizing ASTs and using MinHash and LSH indexing for scalable similarity detection. It supports AI-assisted refactoring workflows by providing a MCP server interface that allows agents to search for duplicates directly. The tool helps catch structural duplications that can lead to bugs and supports clustering of transitive duplicates for comprehensive reporting.