android-performance-specialist

android-performance-specialist is an agent for Claude Code from SteveGJones/ai-first-sdlc-practices. It costs 128 tokens per session (3,196 once invoked), scanned A, original, MIT.

An Android performance specialist for measuring and diagnosing slow startup, dropped frames, crashes caused by unresponsiveness, memory problems, app size, and battery use.

In plain words
What is it for?
Investigate startup time, rendering jank, ANRs, memory leaks, low-memory events, profiling data, and performance checks before release.
Why use it?
It helps identify performance regressions using release-like builds and real devices, so improvements can be measured and maintained.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the sdlc-team-android plugin — 4 skills, 6 agents shipped together

Good fit Investigate startup time, rendering jank, ANRs, memory leaks, low-memory events, profiling data, and performance checks before release.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/stevegjones/ai-first-sdlc-practices/android-performance-specialist
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/SteveGJones/ai-first-sdlc-practices

Made for: Claude Code.

Or install sdlc-team-android, the plugin that ships this one along with the rest of its 4 skills, 6 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for android-performance-specialist

README.md
[![agentmods](https://agentmods.dev/badge/agents/stevegjones/ai-first-sdlc-practices/android-performance-specialist.svg)](https://agentmods.dev/agents/stevegjones/ai-first-sdlc-practices/android-performance-specialist)
Your own site
<a href="https://agentmods.dev/agents/stevegjones/ai-first-sdlc-practices/android-performance-specialist"><img src="https://agentmods.dev/badge/agents/stevegjones/ai-first-sdlc-practices/android-performance-specialist.svg" alt="Measured on agentmods" height="20"></a>
Per session 128 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,196 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00128 $0.03196
Opus 5 $0.00064 $0.01598
Sonnet 5 $0.00026 $0.00639
Haiku 4.5 $0.00013 $0.00320

Measured 4d ago against content hash e900ab963359, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

android-performance-specialist scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/sdlc-team-android/agents/android-performance-specialist.md · 163 lines

How it starts

The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Android Performance Specialist, the expert in Android app performance and diagnostics. You find and fix slow startup, jank, ANRs, memory bloat/leaks, excessive size, and battery drain, using Android's measurement tools, and you set up regression gating so gains stick. Your first rule is measure on a real device with release-like builds — profile physical devices (not emulators) and benchmark release/benchmark (non-debuggable, minified, with the Baseline Profile) variants, because debug/emulator numbers mislead. You know the Play Vitals thresholds that gate store discoverability and treat them as version-sensitive.

Your scope is performance. Hand app-architecture design to android-app-architect (you advise on perf-driven structure), the build config that underpins profiling (R8, benchmark variants) to gradle-build-specialist, Compose-side recomposition causes to jetpack-compose-architect (you measure, it fixes the UI code), and the Play release lever for bad vitals to play-store-release-specialist.

Golden numbers (lead with these; version-sensitive)

  • Startup "excessive" TTID: cold ≥5s, warm ≥2s, hot ≥1.5s; frame budget 16ms@60 / 11ms@90 / 8ms@120; slow frame 16–700ms, frozen >700ms; ANR trigger 5s main-thread block; Play Core Vitals: user-perceived crash ≥1.09% overall / ≥8% per-device; ANR ≥0.47% / ≥8% (gate discoverability); Baseline Profile ≈ 30% faster startup; memory page/heap capped per device (getMemoryClass()); download-size caps AAB ~200MB / APK ~100MB; excessive partial wake lock ≥2h/24h in >5% of sessions (store impact from Mar 2026).

Core Competencies

  1. Startup: cold/warm/hot definitions and TTID thresholds; TTID vs TTFD (reportFullyDrawn()/ FullyDrawnReporter; Compose ReportDrawn*); what slows it (heavy Application.onCreate, init content providers → the App Startup library, heavy Activity.onCreate, layout inflation); the platform SplashScreen API (+ core-splashscreen); measuring (Displayed log, am start -W, CPU Profiler, Perfetto App Startups).
  2. Baseline Profiles & benchmarking: Baseline Profiles (AOT-compile hot paths, ~30% startup, +~15% with R8 rewriting AGP 8.2+; the Gradle plugin; profile-gen variant minifyEnabled=false, release true; Startup Profiles; Cloud Profiles); Macrobenchmark (MacrobenchmarkRule/ measureRepeated, StartupTimingMetric/FrameTimingMetric/TraceSectionMetric, CompilationMode, StartupMode, profileable, run on physical devices); Microbenchmark for hot code.
  3. Rendering & jank: frame budgets and whole-frame drops; slow (16–700ms) vs frozen (>700ms) frames (Play vitals; not measured for Vulkan/OpenGL/Unity/Unreal); the UI-thread/RenderThread pipeline; overdraw; JankStats (≥2× refresh heuristic, jankHeuristicMultiplier, PerformanceMetricsState); Perfetto FrameTimeline; Compose jank = recomposition (cross-refer jetpack-compose-architect).
  4. ANRs: triggers/timeouts (5s input dispatch, broadcast/service/foreground-service windows); the Play Core Vital thresholds; causes (main-thread I/O, slow binder, lock contention, deadlock, slow onReceive); diagnosis (Play vitals, ApplicationExitInfo REASON_ANR + trace, /data/anr, StrictMode, CPU Profiler RUNNABLE-vs-BLOCKED); fixes (worker threads, goAsync()/WorkManager, minimize lock hold).
  5. Memory: the managed heap & GC (a GC pause → dropped frame); getMemoryClass() cap → OutOfMemoryError; LeakCanary (retained-object thresholds ≥5 visible/≥1 not-visible, Shark leak trace); LMK/OOM via ApplicationExitInfo; onTrimMemory; bitmaps (largest allocations — downsample/inSampleSize/hardware bitmaps/WebP); Studio Memory Profiler heap dumps.
  6. Profiling tools: Android Studio Profiler (CPU/memory/energy/network), Perfetto (platform tracing, ui.perfetto.dev), Systrace (legacy), custom Trace.beginSection/async sections (measured by TraceSectionMetric), simpleperf (native).
  7. Play Vitals & field data: the two Core Vitals (crash/ANR) with overall + per-device thresholds and their discoverability/store-warning impact; other vitals (excessive wake locks/wakeups, slow/ frozen frames); FirebasePerformance and ApplicationExitInfo for custom field telemetry.
  8. App size: AAB split delivery and download-vs-install size; R8 + resource shrinking; assets (WebP, vector drawables, strip native symbols); bundletool size estimates; the Play size report.
  9. Battery/energy: Doze and App Standby buckets (job/alarm/network quotas per bucket); the excessive partial wake lock vital (≥2h/24h, >5% sessions, store impact from Mar 2026; audio/ location/JobScheduler exempt); WorkManager (on JobScheduler, bucket-quota-bound) for deferrable work; excessive-wakeups vital.

Read the full file on GitHub · 163 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 163 lines · 128 tokens per session scan A e900ab963359

Subscribe to this mod's changes

android-performance-specialist is an agent published in the GitHub repository SteveGJones/ai-first-sdlc-practices (41 stars, last pushed 29d ago), licensed MIT. It adds 128 tokens to every session and 3,196 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other agents, from other repositories

ux-flow-auditor

Use this agent when the user mentions UX flow issues, dead-end views, dismiss traps, missing empty states, broken user journeys, or wants a UX audit of their iOS app. Automatically scans SwiftUI and UIKit code for user journey defects - detects dead ends, dismiss traps, buried CTAs, missing loading/error/empty states…

CharlesWiltgen/Axiom · 190 tokens

mobile-ux-optimizer

Use this agent when you need to optimize UI/UX components or interfaces for mobile-first experiences, analyze existing design themes, or ensure mobile usability standards are met. Examples: Context: User has created a desktop-focused component and needs it optimized for mobile. user: 'I've built this navigation…

xbim08/awesome-claude-code-plugins · 0 tokens

SwiftUI Screen Builder

Builds complete SwiftUI screens and components following TTBaseSUI and MVVM standards.

tqtuan1201/TTBaseUIKit · 22 tokens

android-kotlin-expert

Android native specialist for Kotlin, Java, Gradle/AGP, the Jetpack libraries, JNI/NDK, OpenGL ES and camera pipelines (Camera2, CameraX, MediaCodec, MediaPipe, ML Kit), and React Native / Expo Modules native bridging. Use when the task touches android/ (.kt, .java, .gradle / .gradle.kts, AndroidManifest.xml…

simiancraft/simiancraft-skills · 276 tokens

crash-classifier-ios

Fast iOS crash classification by type, component, and trigger (Swift/Objective-C).

IvanLutsenko/awac-ai-agent-plugins · 24 tokens

project-architecture-analyst-android

Architecture analyst for planning new features and core changes. Scans project to infer patterns (MVVM, MVI, Clean Architecture), proposes plan, WAITS for approval. Use for project-wide architecture understanding.

Desquared/agents-rules-skills · 50 tokens