# Fuzzloom > Property-based testing and fuzzing for the Fe language, with EVM (sonatina) and native differential testing. Write a property as a Fe function returning bool; Fuzzloom generates inputs inside Fe, shrinks failures, saves them for replay and compares backends. Install: npx fuzzloom. Requires fe 26.6 and Node 22.18. Key facts: properties are `pub fn prop_(args) -> bool`, differential targets `pub fn diff_(args) -> T`; arguments are 1 to 3 of u8..u256, i8..i256 or bool; effects come from `pub fn fixture_*() -> T`; ranges from `/// fuzzloom: in ..=`. Exit codes: 0 passed, 1 finding, 2 infrastructure error. Fuzzing does not prove a property holds. ## Getting started - [Introduction](https://fuzzloom.pages.dev/docs/index.html): Fuzzloom is property-based testing and fuzzing for the Fe language, with differential testing between Fe's two backends. - [Quick start](https://fuzzloom.pages.dev/docs/quickstart.html): From nothing to a shrunk counterexample in five steps. - [Installation](https://fuzzloom.pages.dev/docs/installation.html): doctor checks each requirement and prints a fix for anything missing. It exits 2 if a required piece is missing. A missing native backend is only a warning ## Guides - [Writing properties](https://fuzzloom.pages.dev/docs/writing-properties.html): /// The fee matches exact arithmetic for every amount. pub fn prop_fee_matches_exact_math(_ amount: u64) -> bool { let exact: u128 = (amount as u128) 30 / - [Reading the results](https://fuzzloom.pages.dev/docs/results.html): A shrink that ran out of budget says (budget reached, may not be minimal). The input shown still fails; it may just not be the smallest. - [Effects and fixtures](https://fuzzloom.pages.dev/docs/effects.html): pub fn fixture_ledger() -> Ledger { Ledger { alice: 1000000, bob: 0, fees: 0 } } - [Input ranges](https://fuzzloom.pages.dev/docs/ranges.html): In examples/token-ledger, this annotation turns a vesting overflow that 10,000 uniform inputs never reach into a failure at the first input. - [Differential testing](https://fuzzloom.pages.dev/docs/differential.html): Targets may use effects and declared ranges exactly as properties do. diff needs the native backend; doctor tells you whether it is available. - [Saved failures](https://fuzzloom.pages.dev/docs/saved-failures.html): Every FAIL, ABORT and DIVERGE is saved to .fuzzloom/failures/-.json. The record holds the shrunk input, the original input, the seed, the ite - [Running in CI](https://fuzzloom.pages.dev/docs/ci.html): Exit codes make Fuzzloom a normal CI step: 1 fails the build on a finding, and 2 fails it on an infrastructure problem. Pin a seed for a deterministic chec - [Writing good properties](https://fuzzloom.pages.dev/docs/good-properties.html): - [The whole guide](https://fuzzloom.pages.dev/docs/guide.html): From a first property to a regression suite in CI. For how Fuzzloom works inside, see architecture.md; for exact guarantees, see algorithms.md. ## Reference - [Configuration](https://fuzzloom.pages.dev/docs/configuration.html): Everything is a command-line option or an environment variable. There is no config file. - [Troubleshooting](https://fuzzloom.pages.dev/docs/troubleshooting.html): INFRA ... out-of-gas. fe test caps each EVM test at 10^9 gas. Lower --runs, simplify the property, or run with --backend native, which has no gas. - [CLI reference](https://fuzzloom.pages.dev/docs/cli.html): path is the Fe ingot to work on, a directory with fe.toml and src/. It defaults to the current directory. ## Concepts - [How it works](https://fuzzloom.pages.dev/docs/architecture.html): Fuzzloom has two halves with a strict split. - [Algorithms and guarantees](https://fuzzloom.pages.dev/docs/algorithms.html): Every non-trivial algorithm in Fuzzloom states what it guarantees and where it stops, and has a test that fails if the guarantee breaks. This page collects ## Source - [Repository](https://github.com/MikeMoulder/fuzzloom): the Fe library (ingots/fuzzloom), the CLI (packages/cli), examples and CI - [npm package](https://www.npmjs.com/package/fuzzloom)