# Coq

24 Hacker News commenters recommended Coq and 8 criticized it, November 2025 to November 2026. It was recommended in 9 of 13 months. It meets the Hacker Approved badge rule.

**Verdict:** Free theorem prover people use to write machine-checked proofs, praised for rigor and strong tactics. Caveat: proofs are hard to read and maintain, and adoption is limited.

## Why people recommend it

- Forces careful thinking and strict checking (4 comments): "since the type system is comprehensive enough to force you to actually think about what you really want" ([source](https://news.ycombinator.com/item?id=46560567)) "The proof checker is much more ruthless than the rust compiler." ([source](https://news.ycombinator.com/item?id=46784388)) "The Coq language's rigor really helps with something like this." ([source](https://news.ycombinator.com/item?id=46066012))
- Verifying AI-written code with proofs (3 comments): "ask your agent to verify critical protocols with formal proof in TLA+/lean/coq" ([source](https://news.ycombinator.com/item?id=48106513)) "They are cleaner mathematically, which means AI has a better chance with them." ([source](https://news.ycombinator.com/item?id=48114139)) "Prove your code directly in Lean or Coq or leave it." ([source](https://news.ycombinator.com/item?id=49746888))
- Standard for academic proofs (2 comments): "your academic colleagues will only pay attention to your paper if it comes with a Lean or Coq proof" ([source](https://news.ycombinator.com/item?id=46251806)) "Coq (now Rocq), Isabelle, HOL Light, ACL2 and Mizar that I believe still win" ([source](https://news.ycombinator.com/item?id=49065741))
- Powerful tactics and program extraction (2 comments): "I've found Coq's tactics more powerful and ergonomic" ([source](https://news.ycombinator.com/item?id=47925222)) "Well Coq has program extraction built in." ([source](https://news.ycombinator.com/item?id=46525772))

## What people warn about

- Proofs hard to read and maintain (3 comments): "they seem deliberately designed to require you to use a context explorer to have any hope of understanding the proof at all" ([source](https://news.ycombinator.com/item?id=47240909)) "Most proofs are basically write-only and impossible to maintain" ([source](https://news.ycombinator.com/item?id=49754989)) "I couldn’t make heads or tails of them" ([source](https://news.ycombinator.com/item?id=47930265))
- Awkward syntax and code conversion (3 comments): "Using LEAN or Coq requires you to basically convert your code to LEAN/Coq before you can start proving anything" ([source](https://news.ycombinator.com/item?id=47117373)) "It's a bit of an awkward syntax to get a reliable assembler." ([source](https://news.ycombinator.com/item?id=46066801)) "it feels a little harder than it should be" ([source](https://news.ycombinator.com/item?id=47812205))

## Alternatives hackers recommend

- [Lean 4](https://hackerapproved.com/p/lean.md): 268 recommend
- [Rocq](https://hackerapproved.com/p/rocq.md): 25 recommend
- [Agda](https://hackerapproved.com/p/agda.md): 19 recommend
- [Isabelle](https://hackerapproved.com/p/isabelle.md): 14 recommend
- [Metamath](https://hackerapproved.com/p/metamath.md): 14 recommend

## Cite as

Hacker Approved, "Coq", https://hackerapproved.com/p/coq/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).