litexlang/golitex

Litex: write math, get formalized code. Because the math already is.

What it solves

Litex is a formal language designed to make writing and verifying mathematical proofs more accessible to non-experts. It aims to bridge the gap between how humans naturally construct mathematical arguments and the same rigorous checkable knowledge that is typically found in professional formal verification tools like Lean.

How it works

Litex uses set theory (ZFC) as its mathematical foundation. It allows users to write mathematical facts, definitions, and claims in a Python-like syntax that the Litex verifier checks. As each statement is executed, the verifier searches for local grounds to support the claim; if successful, the accepted facts are added to the current mathematical context, which then serves as the foundation for subsequent steps.

Who it’s for

It is intended for anyone who wants to write checkable mathematics, from students to mathematicians, who may find traditional formal verification languages too steep a learning curve.

Highlights

  • Fact-oriented interface: Users write mathematical statements directly, and the system checks them and reports the ground for the verification.
  • Multilingual feedback: The CLI provides verification feedback in multiple languages (e.g., Chinese, French) to help users understand why a check failed or succeeded.
  • Symbolic input: Supports Unicode mathematical symbols (e.g., ∩, ⊆) for a more natural mathematical writing experience.
  • LaTeX export: Can compile parsed source into mathematical LaTeX documents with prose.
  • AI-assisted workflow: Designed to work with AI by allowing AI to propose source fragments that the verifier can then rigorously check and validate.

Related

  • Project
    lean-dojo/LeanCopilot

    Lean Copilot is an open‑source Lean 4 library that lets you call LLMs from within Lean to suggest tactics, search for proofs, and retrieve useful lemmas. It ships with small pre‑trained models (via CTranslate2) and a plug‑in system for any custom model, enabling neural‑assisted theorem proving directly in the Lean environment.

  • Project
    cameronfreer/lean4-skills

    A workflow pack for AI coding agents that provides structured processes for formal verification and theorem proving in Lean 4.

  • Dispatch
    Lathe: Using LLMs for Hands-On Technical Learning

    Lathe is an experimental tool that uses LLMs to generate custom, multi-part technical tutorials designed for manual implementation, helping developers learn new domains rather than skipping the work.

  • Project
    KKKKhazix/human-writing

    A framework and toolset for AI agents to eliminate "AI-style" Chinese writing by enforcing human-like rhythms, factual substance, and structural checks.