Ataraxy-Labs/sem

Semantic version control => entity-level diffs, blame, and impact analysis on top of git. 28 languages via tree-sitter. Built for coding agents.

What it solves

sem is a semantic version control tool that replaces line-based diffs with entity-level diffs. Instead of seeing which lines changed, it identifies exactly which functions, classes, and methods were modified, added, or removed, making code reviews and change tracking more intuitive for both humans and AI agents.

How it works

It operates as a layer on top of Git, using tree-sitter to parse source code and extract entities. It builds an on-disk query index of the codebase to track dependencies and references. This allows it to perform semantic analysis such as identifying callers and callees, calculating the "impact" of a change (which tests or dependencies are affected), and generating "certificates" for commit ranges that summarize structural changes.

Who it’s for

  • AI Agents: Provides a structured, JSON-compatible interface (including MCP support) to help agents navigate, understand, and verify changes in a codebase.
  • Developers: Offers a more meaningful way to review code changes and avoid the-line-by-line slog of traditional diffs.
  • CI/CD Pipelines: Enables smarter test selection by only running tests that are semantically affected by a change.

Highlights

  • Entity-Level Diffing: Focuses on functions and classes rather than lines of text.
  • Impact Analysis: Identifies exactly which tests and dependent entities are affected by a specific change.
  • sem check: A verification tool that runs compilers, linters, and tests, rechecking only what is provably affected.
  • MCP Support: Serves its capabilities as tools for agents that speak the Model Context Protocol.
  • Promises: Allows users to define code-shape laws via tree-sitter queries to ensure architectural constraints are maintained.

Related

  • Dispatch
    Sem: Semantic Code Understanding via Git Entities

    Sem is a CLI tool that provides semantic, entity-level understanding of Git repositories, improving AI agent accuracy by 2.3x over raw line diffs.

  • Project
    tanishqkancharla/calldiff

    A tool that diffs call stacks across git commits to visualize changes in function call flows, supporting 23 languages and providing structured output for AI agents.

  • Project
    Ataraxy-Labs/weave

    weave is a Rust‑based Git merge driver that parses code with tree‑sitter and merges at the level of functions, classes, and other language entities, automatically resolving false conflicts that arise when multiple agents edit the same file.

  • Project
    alibaba/open-code-review

    An AI-powered code review CLI tool that combines deterministic engineering with an AI agent to provide high-precision, token-efficient reviews of Git diffs and full files.

  • Project
    shuxueshuxue/Spexcode

    A Git-backed spec tree and session manager for coding agents that prevents documentation drift by anchoring specs to code and blocking changes that break the spec-code link.