Ataraxy-Labs/sem

What it solves

sem is a semantic version control tool built on top of Git that replaces line-based diffs with entity-level changes. Instead of seeing which lines changed, users see which functions, classes, or methods were modified, renamed, or moved, making it easier to understand the actual logic changes in a codebase.

How it works

The tool uses tree-sitter to parse code across 32 programming languages and structured data formats. It extracts entities (like functions and classes) and compares them using a three-phase matching process: exact ID matches, structural hashing (which ignores whitespace and comments to detect renames/moves), and fuzzy similarity based on token overlap.

It maintains a SQLite entity cache outside the repository to keep performance high. For very large monorepos, it offers an optional cloud acceleration service to serve pre-built graphs of the codebase.

Who it’s for

• Developers who want more meaningful diffs and impact analysis of their changes.
• AI Coding Agents that need precise, token-efficient context (via an MCP server) rather than reading entire files.
• DevOps/CI pipelines that require entity-level change tracking for pull requests.

Highlights

• Entity-level diffs: Identifies modifications to specific functions, classes, and methods.
• Impact Analysis: A cross-file dependency graph that shows what breaks if a specific entity is changed.
• MCP Server: Integrates with AI agents (like Claude Code or Cursor) to provide tools for impact, context, and diff queries.
• Broad Language Support: Supports 32 languages including Rust, Python, Go, TypeScript, and Java.
• Git Integration: Can be set as the default git diff output via sem setup.
• GitHub Action: Automatically adds entity-level change summaries to pull requests.