Ataraxy-Labs/weave

Entity-level git merge driver. Resolves false conflicts git invents when independent agents edit the same file. ~95% reduction vs. line-based merge.

What it solves

Git's line-based merge often creates "false conflicts" when two developers (or AI agents) add independent code—such as different functions—to the same file. Weave replaces this with entity-level merging, ensuring that changes to different functions, classes, or JSON keys are auto-resolved without requiring manual human intervention.

How it works

Weave uses tree-sitter to parse code into semantic entities (functions, classes, etc.) across three versions of a file: the base, the current branch (ours), and the incoming branch (theirs). It matches these entities by identity (name, type, and scope) and merges them individually. If two branches modify different entities, they are auto-resolved; if they modify the same entity, Weave attempts an intra-entity merge or flags a semantic conflict with detailed context.

Who it’s for

Software developers and AI agent frameworks that frequently perform concurrent edits on the same codebase and want to reduce merge conflict overhead.

Highlights

  • Entity-level granularity: Merges functions and classes rather than lines, resolving conflicts that Git cannot.
  • Broad language support: Supports a wide array of languages including TypeScript, Python, Go, Rust, Java, C++, and more.
  • MCP Server integration: Provides 22 tools for agent frameworks to perform merge analysis, entity inspection, and live coordination via a CRDT.
  • Deterministic merge: Operates as a stateless 3-way merge driver compatible with Git and Jujutsu (jj).
  • Detailed conflict markers: Provides specific reasons why a merge was refused (e.g., which internal guard failed) and the entity involved.

Related

  • Project
    Ataraxy-Labs/sem

    A semantic version control tool built on Git that tracks changes at the entity level (functions, classes, methods) rather than lines, providing deeper insights for humans and AI agents.

  • Project
    wandb/weave

    A toolkit for developing Generative AI applications that provides tracing, logging, and evaluation tools to bring rigor to the LLM development process.

  • Project
    J3n5en/EnsoAI

    EnsoAI is an Electron‑based desktop workbench that couples Git worktree management with persistent AI‑agent sessions (Claude, Gemini, Codex, etc.). It provides a built‑in Monaco editor, visual Git UI, 3‑way merge tool, and one‑click IDE bridging, enabling developers to work on multiple branches in parallel while each branch enjoys its own dedicated LLM context.

  • 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
    Tencent/teamai-cli

    TeamAI CLI is an npm‑distributed command‑line tool that lets a software team keep AI‑assistant resources (skills, rules, docs, hooks, MCP configs, and a shared knowledge base) in a single Git repo and automatically sync them into local Claude, Codex, Cursor, and other agents. It supports role‑/tag‑based filtering, source subscription, session‑learning capture, a code‑base knowledge graph, and dashboards for usage/knowledge‑base health.