Leash.

AI usage

Leash was built with Claude Code (Claude Sonnet 5.5) working alongside the author, over three days.

What the AI did

What the author decided

Where the AI was wrong, and how it was caught

What the AI did for v2

How the work is verified

The AI's claims are not trusted. They are checked by 21 contract tests (including a fuzz test), 10 compiler tests, a local end-to-end drill, a mutation check that breaks the contract three ways and confirms the drill fails each time, and a run on the real testnet with a genuine ML-DSA-87 signature.

Demo video

No model at runtime

The plain-English front end is a small deterministic grammar, not a language model. A model may help a person draft the sentence, but the compiler decides: anything it cannot parse exactly is refused.

View this page on GitHub · raw markdown