aallan

aallan / vera

Public

Vera: a programming language designed for LLMs to write

veralang.dev algebraic-effects contracts formal-verification llm parser
24
2
100% credibility
Found Mar 03, 2026 at 24 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Vera is a programming language designed specifically for AI models to generate verifiable code with mandatory contracts, explicit effects, and no variable names.

How It Works

1
💡 Discover Vera

You learn about Vera, a special language that lets AI create safe, reliable programs with built-in promises.

2
🛠️ Set up Vera

You easily download and prepare Vera on your computer so it's ready to use.

3
✏️ Make a program

You write a simple program or ask an AI to generate one, adding clear promises about what it does.

4
✅ Check your work

You quickly review the program to ensure it follows all the safety rules perfectly.

5
🔍 Verify promises

Vera automatically confirms that your program's promises are true and reliable.

6
▶️ Run it safely

You launch the program and watch it work exactly as promised, with full confidence.

🎉 Success!

Your AI-generated program runs flawlessly, verified and secure every time.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 24 to 24 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is vera?

Vera is a programming language designed for LLMs to generate code, solving their struggles with coherence across large programs by mandating explicit contracts, algebraic effects, and typed positional references (@T.n) instead of variable names. It compiles to WebAssembly for sandboxed execution, with CLI tools like `vera check`, `vera verify` (using Z3 for proofs), and `vera run` to test pure functions or IO effects. Python-based, it produces machine-checkable output where errors come as natural-language fixes for LLMs.

Why is it gaining traction?

Its LLM-first design turns verification into a strength—contracts prove preconditions/postconditions statically, effects track state explicitly, and diagnostics guide iterative fixes, unlike fuzzy code in Python or Rust. WASM portability and a rigid canonical format minimize syntax debates, while features like refinement types catch errors early. Detailed spec docs and agent skills files make it plug-and-play for AI code gen.

Who should use this?

LLM engineers building agentic workflows that output executable code, researchers experimenting with verified functions for math or data pipelines, or Python devs exploring algebraic effects for safer state management in prototypes.

Verdict

With 24 stars and 1.0% credibility score, Vera's pre-alpha status shows in limited ecosystem but impresses with 88% test coverage, 13-chapter spec, and working examples—grab it for LLM programming language experiments if you're okay with active development.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.