Dejavu Onboarding
Communitymttmcknn
Install DejaVu and prove recomposition tracking in an Android or Compose Multiplatform project that does not have it yet. Use for dependency, compatibility, compiler source information, test host, or first-test setup. For an already configured project, use test writing or failure triage instead.
Overview
- Repository
- mttmcknn/dejavu
- Publisher
- mttmcknn
- Stars
- 197
- Forks
- 4
- Bundled files
- 1
- License
- Apache-2.0
- Updated
- Nov 8, 2026
Installation
tmp=$(mktemp -d)
git clone --depth 1 https://github.com/mttmcknn/dejavu.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/dejavu-onboarding" ~/.claude/skills/dejavu-onboarding
rm -rf "$tmp"tmp=$(mktemp -d)
git clone --depth 1 https://github.com/mttmcknn/dejavu.git "$tmp"
mkdir -p ~/.codex/skills
cp -R "$tmp/skills/dejavu-onboarding" ~/.codex/skills/dejavu-onboarding
rm -rf "$tmp"Install the Dejavu Onboarding agent skill from https://github.com/mttmcknn/dejavu/tree/main/skills/dejavu-onboarding into this project's skills folder.Instructions
DejaVu onboarding
Inspect the existing version catalog, plugins, SDK levels, source sets and test host. Add a compatible test dependency and one test that proves tracking is active. Preserve dependency conventions and the user's version constraints.
Choose the compatible setup
Read setup.md for the versioned dependency contract and Android/KMP requirements. It is bundled so installation outside the DejaVu repository works. Use the host project's files, not assumed module paths. For newer versions, consult their release-specific documentation before selecting one; don't silently downgrade or upgrade Compose, Kotlin or SDK levels to fit an example.
If prerequisites conflict, identify the exact mismatch and compatible DejaVu line. Continue useful diagnosis; request a version decision only when the task has not already authorized it. If setup is complete, leave it alone.
Prove tracking works
- Android: retain the activity host with
createRecompositionTrackingRule<YourActivity>(), or use the plain factory for test-owned content. Include Compose test APIs explicitly and the debug test manifest when needed. Do not nest independent Compose test environments. - KMP: use
runRecompositionTrackingUiTest { setTrackedContent { … } }and return itsTestResultfrom the test method. Assertions and cleanup stay inside the suspending body so Wasm observes failures. - Use a tagged user composable and a real state-changing interaction. Settle initial content, reset counts preserving live composition history, perform the interaction, settle, and assert the changed UI and a nonzero count. A stable-only assertion can pass when tracking is broken.
- Preserve existing UI assertions. Library accuracy fixtures need an independent
unkeyed
SideEffectcounter and expected budget failures. Intentionally excessive recompositions are the behavior under test.
Run the narrowest relevant actual UI test available. Report the dependency line, source set, host and observed result. Compilation does not prove tracking; report a missing device or runtime as an unrun check.
For more coverage, dejavu-test-writer is useful when available; it is not a
prerequisite to finishing. Public reference:
versioned setup.
Bundled files
FAQ
What does Dejavu Onboarding do?
Install DejaVu and prove recomposition tracking in an Android or Compose Multiplatform project that does not have it yet. Use for dependency, compatibility, compiler source information, test host, or first-test setup. For an already configured project, use test writing or failure triage instead.
Who made Dejavu Onboarding?
mttmcknn (a community member) publishes it in the mttmcknn/dejavu repository on GitHub.
Is Dejavu Onboarding free?
Yes. Dejavu Onboarding is open source, published by mttmcknn under the Apache-2.0 license.
What comes with Dejavu Onboarding?
Its instructions and 1 file the agent reads only when a task needs them, such as references, examples or scripts.
Which agents can use Dejavu Onboarding?
Any coding agent that reads SKILL.md files, including Claude Code, Codex, Cursor, GitHub Copilot and Gemini CLI. The agent loads the skill when a task matches its description.
What are agent skills?
An agent skill is a folder with a SKILL.md file: instructions that teach a coding agent how to do one kind of task, and optionally files it can read while it works. The agent keeps only the skill's name and description in mind and reads the rest when a task calls for it.
How are agent skills different from MCP servers?
An MCP server gives an agent new tools to call, code that runs somewhere and returns results. A skill gives it knowledge: how to approach a task with the tools it already has. Many agents use both.
How can my agent test my Android app?
Robomori lets coding agents create, start and control Android emulators and devices from one command, so they can install your app, take screenshots and check dark mode, font sizes and languages.
Analyze and create a testing strategy for Android apps - install testing libraries, set up test infrastructure, create harnesses for unit tests, UI tests, screenshot tests, and end-to-end tests.
Write, review, migrate, and debug Kotlin tests with Kotest 6.x: spec styles (FunSpec, DescribeSpec, BehaviorSpec, StringSpec, ...), matchers (shouldBe, shouldThrow, assertSoftly, eventually), data-driven tests (withTests/withData), property-based testing (Arb, checkAll, forAll, permutations), lifecycle hooks, extensions (Spring, Testcontainers, Koin), ProjectConfig, coroutine/virtual-time testing, tags, and Kotlin Multiplatform setup. Use when a project depends on io.kotest, when editing *Spec/*Test.kt files that import io.kotest, when the user mentions Kotest, kotest matchers, Arb or property testing in Kotlin, or when migrating from JUnit 4/5, TestNG, or Spek to Kotest.
Use when writing or reviewing Jetpack Compose UI tests, screenshot tests or baseline-recording evidence, previews, semantics assertions, fake image loading, keyboard input, focus assertions, interaction state (hover/pressed/focused), or tests for plain state-driven UI composables.
Use when working with Roborazzi screenshot tests on Android/JVM — setting up the Roborazzi Gradle plugin, running record/compare/verify tasks, writing tests with captureRoboImage or RoborazziRule, Compose Preview screenshot testing (ComposablePreviewScanner), Compose Multiplatform (iOS/desktop) screenshots, AI-powered image assertions, roborazzi.* Gradle properties, or troubleshooting Roborazzi/Robolectric screenshot test failures.
