vercel-labs/zerolang

The Programming Language for Agents

What it solves

Zerolang addresses the inefficiency of AI agents coding via text. Traditional agent loops rely on writing and editing text files, which often leads to guessing, formatting errors, and a cycle of trial-and-error. Zerolang replaces text as the source of truth with a semantic graph, allowing agents to interact directly with the compiler's internal representation.

How it works

Instead of editing line ranges in a text file, agents use a graph-native approach. They query the program database (zero.graph) for symbols, types, and ownership facts, and then submit "checked patches" to modify the semantic structure. The compiler validates these edits before they are written to the store. While the graph is the primary authoring surface for agents, humans can still interact with the code through .0 files, which are human-readable projections of the graph.

Who it’s for

It is designed for AI agents and developers building agentic workflows who want to reduce token usage and increase the reliability of automated code generation and modification.

Highlights

  • Graph-Native Database: Uses a semantic graph as the program database rather than source text.
  • Checked Edits: Agents submit patches that are validated by the compiler to prevent stale or invalid changes.
  • Projection Boundary: Explicit import/export mechanisms allow humans to review or edit text projections without silently diverging from the graph.
  • Agent-Friendly Tooling: Provides specific commands like zero query and zero patch to expose compiler facts directly to agents.

Related

  • Project
    vercel-labs/lat.md

    lat.md is an npm‑distributed CLI that creates a markdown‑based knowledge graph for a codebase. It lets you write linked documentation in `lat.md/`, annotate source files with `@lat` comments, and validates the graph. Agents and developers can search it semantically (offline MiniLM or remote OpenAI/Vercel embeddings) and use commands like `lat check`, `lat locate`, and `lat expand` to keep design decisions, constraints, and test specs in sync with the code.

  • Project
    trailhq/Graft

    Graft is an open‑source TypeScript/Node.js tool that builds a local, markdown‑based knowledge graph of a codebase (summaries, key snippets, typed links) using tree‑sitter for structural parsing and optional LLM summarisation. The graph is cached in a `graft/` folder, refreshed only on file changes, and automatically wired into coding agents like Claude Code, Cursor, Codex, and Gemini. Benchmarks claim up to 46 % fewer tool calls, 42 % token savings, 60 % latency reduction, and a 12‑point correctness boost on SWE‑bench. Integration is via a single `graft init` command that writes agent‑specific instruction files; the CLI also offers search, visualization, and telemetry controls. The project is MIT‑licensed and targets faster, cheaper AI‑assisted development.

  • Project
    colbymchenry/codegraph

    CodeGraph is a Rust‑based local tool that builds a full‑code knowledge graph for any project (supports dozens of languages) and integrates it with AI coding assistants (Claude Code, Cursor, Copilot, Gemini, etc.). The graph lets agents retrieve exact symbols and call‑paths in a single request, cutting tool calls, file reads, token usage and cost dramatically. Install via a one‑line script, run `codegraph init` per repo, and the CLI auto‑updates the graph as you edit. A browser UI (`codegraph ui`) visualises the graph, and the CLI can uninstall itself cleanly.

  • Project
    getnao/nao

    An open-source framework to build and deploy analytics agents that enable business users to query data using natural language and generate visualizations.

  • Project
    xoai/sage-wiki

    A graph-based knowledge base that uses LLMs to compile raw documents into an interlinked wiki and knowledge graph for shared human-AI memory.