Android Profiler
Android (Google)OfficialCreated by Google LLCDistributed by Android (Google)
- Skill
- Tool integration
- Software development
- Apache-2.0
- Low risk
- Has scripts
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
The android-profiler skill provides a structured approach to capturing and interpreting Android performance data. It acts as an orchestrator for recording system traces, heap dumps, and CPU profiles, and provides workflows for deep-dive analysis using Perfetto. The skill is designed to help developers identify the root causes of UI jank, memory leaks, battery drain, and system-level performance regressions by leveraging standardized trace analysis and SQL-based querying.
What it does
- Records system traces, CPU call stacks, and memory heap profiles.
- Analyzes Perfetto traces to identify UI jank, frame drops, and startup latency.
- Investigates system-level issues like IPC Binder storms and thread scheduling stalls.
- Diagnoses power-related issues including battery drain and kernel wake locks.
- Debugs memory pressure, LMK events, and native or Java heap leaks.
- Executes custom PerfettoSQL queries for advanced trace data exploration.
When to use it
- When you need to investigate app startup latency or intermittent UI jank.
- When diagnosing memory-related crashes like OOMs or excessive allocation churn.
- When analyzing system-wide performance regressions using A/B trace comparisons.
- When you need to profile power consumption or modem power rail activity.
- When you need to perform ad hoc analysis on complex Android system traces.
Advantages
- Provides a unified workflow for both recording and analyzing performance data.
- Reduces ambiguity by enforcing clear intent disambiguation before execution.
- Includes specialized reference hints for CPU, memory, graphics, and power analysis.
- Supports advanced data exploration via PerfettoSQL integration.
- Ensures consistent environment setup through standardized path management.
Usage examples
- “My app is stuttering during scrolling. Can you help me find the cause?”
- The agent will initiate a trace recording session focused on UI jank, then guide you through analyzing the resulting Perfetto trace for frame drops and thread stalls.
- “I suspect a memory leak in my application. How can I investigate?”
- The agent will help you capture a Java or native heap dump and provide steps to analyze the memory allocation patterns to identify potential leaks.
Install
Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/android/android-profiler/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip android-profiler.zip -d .agents/skills
Documentation by Google LLC
Your primary role is to figure out what the user wants to do (disambiguation), find the right instructions for it (workflow discovery), and follow them step-by-step (following the execution plan). If the request is unclear, work with the user to finalize the execution plan before proceeding.
Prerequisites and Setup
Before executing any workflows, read references/env_setup.md. It defines
what to set $SKILL_ROOT to - the anchor every other path in this skill is
written against.
Intent Disambiguation
Do not guess the user's intent. If the user request is not clear, ask the user what they want to do before proceeding.
Recording
Route all recording requests through recording/recording_orchestrator. This defines guidelines and pre-flight checks or dependency checks that apply to all recording workflows, and ensures you have the necessary setup to proceed. Read the orchestrator and execute the plan it describes based on what the user wants to record (for example, a system trace or a heap dump).
Analysis
Route all analysis requests through analysis/analysis_orchestrator.
Files
- ATTRIBUTION.mdAdded by AutoGineer965 B
- LICENSELicence file added by AutoGineer11.1 KiB
- SKILL.md2.2 KiB
- analysis/analysis_orchestrator.md1.5 KiB
- analysis/workflows/perfetto-sql/perfetto_sql.md969 B
- analysis/workflows/perfetto-trace-analysis/perfetto_trace_analysis.md4.2 KiB
- analysis/workflows/perfetto-trace-analysis/references/hints/hints_cpu.md2.6 KiB
- analysis/workflows/perfetto-trace-analysis/references/hints/hints_graphics.md2.1 KiB
- analysis/workflows/perfetto-trace-analysis/references/hints/hints_io.md1.8 KiB
- analysis/workflows/perfetto-trace-analysis/references/hints/hints_ipc.md1.7 KiB
- analysis/workflows/perfetto-trace-analysis/references/hints/hints_memory.md2.2 KiB
- analysis/workflows/perfetto-trace-analysis/references/hints/hints_power.md1.2 KiB
- analysis/workflows/perfetto-trace-analysis/references/per_candidate_analysis.md8.8 KiB
- analysis/workflows/perfetto-trace-analysis/references/triage.md3.9 KiB
- recording/recording_orchestrator.md1.5 KiB
- recording/workflows/perfetto-trace-recording/perfetto_trace_recording.md4.7 KiB
- recording/workflows/perfetto-trace-recording/references/example-configs/app_jank.pftxt1.0 KiB
- recording/workflows/perfetto-trace-recording/references/example-configs/cpu_profile.pftxt773 B
- recording/workflows/perfetto-trace-recording/references/example-configs/java_heap_dump.pftxt566 B
- recording/workflows/perfetto-trace-recording/references/example-configs/long_background.pftxt1006 B
- recording/workflows/perfetto-trace-recording/references/example-configs/memory_counters.pftxt1.0 KiB
- recording/workflows/perfetto-trace-recording/references/example-configs/native_heap.pftxt589 B
- recording/workflows/perfetto-trace-recording/references/example-configs/sched_cpu.pftxt836 B
- recording/workflows/perfetto-trace-recording/references/trace_config_reference.md4.9 KiB
- references/env_setup.md513 B
- references/perfetto/bin/trace_processorScript10.8 KiB
- references/perfetto/setup.md1.3 KiB
- references/perfetto/sql.md15.8 KiB
analysis/workflows/perfetto-trace-analysis/references/hints/hints_memory.md
Expert hints for memory pressure, Low Memory Killer (LMK), and allocations. keywords:
memory
lmk
lmkd
oom
rss
swap
kswapd
thrashing
pressure
- allocation
To investigate low memory kills, look for the
lmk_kill_occurredftrace event; a high count indicates severe memory pressure.When you see a burst of
lmk_kill_occurredevents, query for processes with high CPUwall_durationin thesched_slicetable during the same time window. Runaway processes consuming CPU exacerbate memory pressure.To understand a process's memory impact, inspect its
anon_rss(Anonymous Resident Set Size) from memory counters likemem.info.Check system-wide memory stats for a high or rapidly increasing
swap_usedvalue.To confirm memory thrashing, look for high CPU usage by the
kswapdkernel thread.To find the direct trigger for LMKD, look for
memory_pressuretrace events from Pressure Stall Information (PSI).To detect 'lost' memory for processes using hardware accelerators (like a TPU), query the
countertable forRssFilevalues. If it drops significantly while hardware is active, check forkswapd0scheduling slices.When analyzing memory shared with hardware, query
dma_heap_statanddmabuf_total_sizecounters (orion_total_sizefor older devices) for a more accurate picture than RSS.Be aware that
RssFilecan over-report memory if the same physical page is mapped multiple times.When investigating OOMs, establish a baseline by querying process memory counters (for example,
mem.rss) and compare the median and 95th percentile against its history.If bitmaps are a major memory consumer, check for outliers in bitmap count by querying
android.graphics.Bitmapinstances across traces.To find what is holding onto an object (like a bitmap), trace its retainer path back to a GC root by querying
heap_graph_reference. Pay close attention to custom application classes.Query the
android_bitmapstable to check the width and height properties of bitmaps.Check for the presence of software bitmaps, which consume memory on both the app heap and in graphics memory.
When analyzing bitmaps, look for duplicates by checking for multiple
android.graphics.Bitmapobjects with identical properties.
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Security
What our scanner noticed· Low risk
- Scripts that your agent may run
This skill includes scripts that your agent may run. Read them before you use it.
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Created by
- Google LLC
- Publisher
- Android (Google) (Official)
- Distributed by
- Android (Google)
- Source
- android/skills · profilers/android-profiler · commit bf68e2d
- Licence
- Apache License 2.0 (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE.txt)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Google LLC or Android (Google).
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download android-profiler.zip