Faster tests you can actually trust.

For teams and coding agents waiting on slow or unreliable test feedback: understand failures, identify flakes, and distribute test work without replacing your CI.

Read docs

Features

Faster tests, fewer surprises

Tuist is build infrastructure for engineering teams and coding agents.

TuistAppleArchiveTests TuistAuthCommandTests TuistAutomationTests TuistCASTests TuistCITests TuistCacheEETests TuistCoreTests TuistGeneratorTests TuistKitTests TuistLoaderTests TuistServerTests TuistSupportTests

41
```
final class AuthSessionStore {
```

    private let keychain: KeychainClient

    func restore() throws -> AuthSession? {

        guard let data = try keychain.read(.session) else {

46
```
            return nil
```
47
```
        }
```

        return try decoder.decode(AuthSession.self, from: data)

49
```
    }
```

        let session = try decoder.decode(AuthSession.self, from: data)

        guard session.expiresAt > Date.now else {

            try keychain.delete(.session)

53
```
            return nil
```
54
```
        }
```
55
```
        return session
```
56
```
    }
```
57

    func persist(_ session: AuthSession) throws {

        let data = try encoder.encode(session)

        try keychain.write(data, for: .session)

61
```
    }
```
62
```
}
```

Run only what changed

In Tuist-generated projects, the default selective-testing strategy skips test targets whose hashed inputs have not changed since their last successful run. It is not line-level impact analysis.

0ms 5ms 10ms 15ms 20ms 25ms 30ms 35ms 40ms 45ms 50ms 55ms

Split tests intelligently

Run large test suites across multiple CI runners without creating one slow shard. Tuist uses historical test timings to balance tests so each runner finishes at roughly the same time.

Test runs 75

Skipped tests 5

Failed test runs 15

p90 test run duration 1m 40s

Know what's slowing you down

Track test performance over time, uncover slow tests, and understand failed runs without digging through CI logs.

Diagnose slow tests

Stop flakes from blocking CI

Detect inconsistent test outcomes and track them over time. Configure quarantine where supported to mute or skip known flakes while someone owns fixing them; quarantine is not a repair.

Handle flaky tests

What are you waiting for?

Supercharge your development

Get started Talk to us

Cache

Compute

Tests

Previews

Generated projects

Registry

About

Public dashboard

Customers

Slack

Newsletter

GitHub

Community forum