AgentRules.org
RulesStacksClientsAuditorPrunerSelf-HealRFCsBenchmarkCLICI/CDAuthorsSubmit
This specification is maintained by the AgentRules Working Group. Have enhancements or want to claim maintainership?Submit Specification →
mobilecritical PriorityID: cmp-skiko-gpu-context-loss-recovery-standard

Compose Multiplatform Skiko GPU Context Loss Recovery & Surface Lifecycle Standard

Enforces deterministic teardown and recreation of Skiko render targets across mobile display lifecycle transitions. Eliminates silent render pipeline freezes, black screens, and native SIGSEGV crashes caused by stale Metal or Vulkan GrDirectContext handles when mobile operating systems reclaim hardware display buffers.

Prohibited Anti-Pattern: Silent Skiko Render Loop Stall via Stale Hardware Context Retention and Unhandled Metal/Vulkan Surface Invalidation

Export Ready-to-Drop Configuration

Select format below to copy code or download file directly.

---
description: Compose Multiplatform Skiko GPU Context Loss Recovery & Surface Lifecycle Standard - Enforces deterministic teardown and recreation of Skiko render targets across mobile display lifecycle transitions. Eliminates silent render pipeline freezes, black screens, and native SIGSEGV crashes caused by stale Metal or Vulkan GrDirectContext handles when mobile operating systems reclaim hardware display buffers.
globs: **/commonMain/**/*.kt, **/iosMain/**/*.kt, **/androidMain/**/*.kt, **/composeApp/**/*.kt
alwaysApply: true
verified_by: "agentrules.org"
license: "CC-BY-4.0"
source: "https://agentrules.org/rules/cmp-skiko-gpu-context-loss-recovery-standard"
---

# Compose Multiplatform Skiko GPU Context Loss Recovery & Surface Lifecycle Standard

> **Severity**: CRITICAL | **Category**: mobile
> **Anti-Pattern Warning**: Silent Skiko Render Loop Stall via Stale Hardware Context Retention and Unhandled Metal/Vulkan Surface Invalidation

## System Constraints
You are a Principal Software Architect enforcing deterministic GPU resource management in Compose Multiplatform applications using Skiko. Mandate strict encapsulation of hardware-backed Skia primitives, deterministic lifecycle coupling with native surfaces (CAMetalLayer and Android SurfaceHolder), and automated fallback and recovery mechanics when the underlying GrDirectContext is invalidated by the operating system compositor.

## Strict Guidelines
- Engineers MUST ALWAYS invalidate, abandon, or purge stale `org.jetbrains.skia.DirectContext` allocations immediately upon receiving native mobile lifecycle suspension signals (`onStop` on Android, `applicationDidEnterBackground` on iOS).
- Engineers MUST NOT cache hardware-backed `org.jetbrains.skia.Image`, `org.jetbrains.skia.Surface`, or `org.jetbrains.skia.Shader` instances in long-lived singletons or across Compose recomposition boundaries without an explicit generation/epoch token.
- Code SHOULD intercept `RenderException` and native surface allocation failures within custom Skiko rendering delegates to trigger a graceful fallback to a fresh swapchain before re-throwing.
- Services MUST re-evaluate `SkiaLayer.needRedraw()` and re-trigger surface configuration when returning from background states to prevent deadlocks in the display link loop.
- Components NEVER hold strong references to `SkiaLayer` or native platform display layers inside unmanaged coroutines or global state handlers.

## Bad vs Good Implementations
### Skia Hardware Context Lifecycle & Surface Invalidation Recovery

#### Bad Antipattern (AI Common Hallucination)
> **Why it fails**: Retaining raw GPU-backed Skia objects without reacting to native OS context teardown causes silent rendering stalls or native crashes. When the OS reclaims the GPU context during backgrounding, the cached Image becomes an invalid texture handle; calling drawImage on a stale context halts the Skiko render loop.
```kotlin
class CustomSkikoRenderer : SkikoView {
    // ANTI-PATTERN: Retaining raw hardware surface and texture across app lifecycle
    private var cachedImage: org.jetbrains.skia.Image? = null
    private var directContext: org.jetbrains.skia.DirectContext? = null

    override fun onRender(canvas: org.jetbrains.skia.Canvas, width: Int, height: Int, nanoTime: Long) {
        if (cachedImage == null) {
            // Allocates once and never verifies if GPU context was dropped by OS
            cachedImage = org.jetbrains.skia.Image.makeFromEncoded(readAssetBytes("hero.png"))
        }
        // Fails silently or crashes with SIGSEGV/Metal device lost when app resumes
        canvas.drawImage(cachedImage!!, 0f, 0f)
    }
}
```

#### Production Standard (Correct)
> **Standard**: Adheres to the recovery standard by hooking into the lifecycle observer to invalidate and close native Skia resources upon suspension. Catching RenderException and nullifying references forces a clean texture upload against the recreated GPU context on the subsequent render cycle, preventing display pipeline freezes.
```kotlin
class CustomSkikoRenderer(
    private val lifecycleOwner: LifecycleOwner
) : SkikoView, DefaultLifecycleObserver {
    private var cachedImage: org.jetbrains.skia.Image? = null
    private var contextEpoch: Long = 0L

    init {
        lifecycleOwner.lifecycle.addObserver(this)
    }

    override fun onStop(owner: LifecycleOwner) {
        // INVARIANT: Explicitly release GPU texture handles on suspension
        releaseGpuResources()
    }

    override fun onRender(canvas: org.jetbrains.skia.Canvas, width: Int, height: Int, nanoTime: Long) {
        try {
            val activeImage = cachedImage ?: createAndCacheTexture(canvas)
            canvas.drawImage(activeImage, 0f, 0f)
        } catch (e: org.jetbrains.skiko.RenderException) {
            // Recover deterministically from context loss
            releaseGpuResources()
            contextEpoch++
        }
    }

    private fun createAndCacheTexture(canvas: org.jetbrains.skia.Canvas): org.jetbrains.skia.Image {
        return org.jetbrains.skia.Image.makeFromEncoded(readAssetBytes("hero.png")).also {
            cachedImage = it
        }
    }

    private fun releaseGpuResources() {
        cachedImage?.close() // Releases Skia native ref-count
        cachedImage = null
    }
}
```
Recommended MCP Servers (mcp-serverlist.com)

Equip agent with system tools to verify code for this stack.

Gradle Environment Inspector
Inspects Multiplatform Skiko dependencies, target configurations, and compile-time flags across iOS and Android modules.
Recommended Agent Skills (askills.dev)

Ready-to-use executable skill functions tailored to this stack.

KMP Environment & Skiko Diagnostic CLI
Validates Kotlin Multiplatform project target linkages, Metal framework dependencies, and Skiko rendering configuration.
Architectural Specification & Invariant Rationale

Architectural Invariant Design: Compose Multiplatform Skiko GPU Context Loss Recovery & Surface Lifecycle Standard

Compose Multiplatform Skiko GPU Context Loss Recovery & Surface Lifecycle Standard defines an industrial-grade behavioral standard for mobile engineering. Enforces deterministic teardown and recreation of Skiko render targets across mobile display lifecycle transitions. Eliminates silent render pipeline freezes, black screens, and native SIGSEGV crashes caused by stale Metal or Vulkan GrDirectContext handles when mobile operating systems reclaim hardware display buffers. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Silent Skiko Render Loop Stall via Stale Hardware Context Retention and Unhandled Metal/Vulkan Surface Invalidation Without an explicit system instruction file, developers spend hours debugging silent failures.
Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure ModeAI Symptom (The Anti-Pattern)SeverityProduction Impact & Remedy
AI Hallucination & Deprecation DriftSilent Skiko Render Loop Stall via Stale Hardware Context Retention and Unhandled Metal/Vulkan Surface InvalidationCritical
Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production.
Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture InconsistencyAI assistants produce mixed paradigms across files, creating fragmented codebase conventions.Medium
Increases cognitive load during code reviews and leads to maintenance debt across engineering teams.
Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.
Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

1

Engineers MUST ALWAYS invalidate, abandon, or purge stale `org.jetbrains.skia.DirectContext` allocations immediately upon receiving native mobile lifecycle suspension signals (`onStop` on Android, `applicationDidEnterBackground` on iOS).

Engineers MUST ALWAYS invalidate, abandon, or purge stale `org.jetbrains.skia.DirectContext` allocations immediately upon receiving native mobile lifecycle suspension signals (`onStop` on Android, `applicationDidEnterBackground` on iOS).

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
2

Engineers MUST NOT cache hardware-backed `org.jetbrains.skia.Image`, `org.jetbrains.skia.Surface`, or `org.jetbrains.skia.Shader` instances in long-lived singletons or across Compose recomposition boundaries without an explicit generation/epoch token.

Engineers MUST NOT cache hardware-backed `org.jetbrains.skia.Image`, `org.jetbrains.skia.Surface`, or `org.jetbrains.skia.Shader` instances in long-lived singletons or across Compose recomposition boundaries without an explicit generation/epoch token.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
3

Code SHOULD intercept `RenderException` and native surface allocation failures within custom Skiko rendering delegates to trigger a graceful fallback to a fresh swapchain before re-throwing.

Code SHOULD intercept `RenderException` and native surface allocation failures within custom Skiko rendering delegates to trigger a graceful fallback to a fresh swapchain before re-throwing.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
4

Services MUST re-evaluate `SkiaLayer.needRedraw()` and re-trigger surface configuration when returning from background states to prevent deadlocks in the display link loop.

Services MUST re-evaluate `SkiaLayer.needRedraw()` and re-trigger surface configuration when returning from background states to prevent deadlocks in the display link loop.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
5

Components NEVER hold strong references to `SkiaLayer` or native platform display layers inside unmanaged coroutines or global state handlers.

Components NEVER hold strong references to `SkiaLayer` or native platform display layers inside unmanaged coroutines or global state handlers.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.
Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.
Technical Questions & Answers

Frequently Asked Technical Questions

Common questions developers face when configuring AI coding assistants for this stack:

Q:Why is Compose Multiplatform Skiko GPU Context Loss Recovery & Surface Lifecycle Standard critical for AI coding assistants?

Enforces deterministic teardown and recreation of Skiko render targets across mobile display lifecycle transitions. Eliminates silent render pipeline freezes, black screens, and native SIGSEGV crashes caused by stale Metal or Vulkan GrDirectContext handles when mobile operating systems reclaim hardware display buffers. Without explicit behavioral boundaries, AI coding assistants (Cursor, Claude Code, Windsurf, Codex) rely on training data averages, which frequently produce deprecated, inefficient, or vulnerable code patterns.

Q:How does adopting this rule prevent AI hallucinations?

This rule explicitly declares what the AI must avoid (Silent Skiko Render Loop Stall via Stale Hardware Context Retention and Unhandled Metal/Vulkan Surface Invalidation) and prescribes verified positive examples. Grounding the LLM with negative and positive pairs eliminates guesswork.

Q:Which AI coding environments are supported?

This specification is multi-platform compatible: Cursor (.cursor/rules/*.mdc & .cursorrules), Anthropic Claude Code (CLAUDE.md), Windsurf (.windsurfrules), Codex / Cline (AGENTS.md), and GitHub Copilot (.github/copilot-instructions.md).

Q:Can I combine this rule with other technology stacks?

Yes. Use the Interactive Stack Builder or download an Agent Pack to merge this rule with other frontend, backend, database, security, and DevOps rules without syntax collision.

Open Specification · CC BY 4.0·Free for Personal, Team & Commercial In-Repo Use

Citation & Open Source Attribution Guidelines

This specification is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). You are completely free to adapt and share this rule in your workspace. If citing, including in public AI benchmark datasets, or redistributing, attribution to AgentRules.org is required.

Related MOBILE Rules

View all rules →
mobile
106

Tauri 2.0 Mobile IPC & Capabilities Permissions Matrix Standard

Tauri v2 introduces a capability-driven permissions system replacing legacy v1 allowlists across iOS and Android runtimes. Failure to constrain IPC command handlers with typed schemas and explicit mobile capability definitions exposes native mobile OS surfaces to arbitrary code execution, permission leakage, and silent serialization failures. Production mobile deployments require strictly typed IPC boundary validation, granular capability scopes, and graceful permission degradation.

mobile
70

Capacitor 7 Native Bridge, Biometric Auth & Android 15 Lifecycle Production Standard

Capacitor 7 enforces strict bridge lifecycle contracts, asynchronous native plugin interop, and mandatory Android 15 edge-to-edge window inset handling. Storing sensitive authentication tokens in unencrypted web storage or failing to bind biometric authentication directly to hardware-backed Keystore or Secure Enclave mechanisms compromises app integrity and causes fatal crash or viewport occlusion issues on modern mobile operating systems.

mobile
58

Tauri 2.0 Mobile IPC Isolation and Capabilities Permission Scoping

Enforces strict capability permission boundaries, payload validation, and secure IPC invoke scoping for Tauri v2 Android and iOS environments. Prevents arbitrary IPC command execution, excessive native privilege granting, and deserialization vulnerabilities across the mobile webview-native boundary.

1 Rule in Stack

Compose