perfetto-mcp

Overview

Perfetto MCP is a specialized Model Context Protocol server that enables developers to analyze Android performance traces using natural language prompts. It converts plain English questions into precise Perfetto SQL queries, allowing users to diagnose ANRs, find CPU bottlenecks, detect UI jank, analyze memory leaks, and investigate binder transactions without writing SQL. The server supports multiple MCP integrations including Cursor, Claude Code, VS Code, and Codex, with both uvx and pip installation options.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

• find_slices

  Survey slice names and locate hot paths

  Try:Find slice names containing 'Choreographer' and show top examples
• execute_sql_query

  Run custom PerfettoSQL for advanced analysis

  Try:Run custom SQL to correlate threads and frames in the first 30s
• detect_anrs

  Find ANR events with severity classification

  Try:Detect ANRs in the first 10s and summarize severity
• anr_root_cause_analyzer

  Deep-dive ANR causes with ranked likelihood

  Try:Analyze ANR root cause around 20,000 ms and rank likely causes
• cpu_utilization_profiler

  Thread-level CPU usage and scheduling

  Try:Profile CPU usage by thread and flag the hottest threads
• main_thread_hotspot_slices

  Find longest-running main thread operations

  Try:List main-thread hotspots >50 ms during 10s–25s
• detect_jank_frames

  Identify frames missing deadlines

  Try:Find janky frames above 16.67 ms and list the worst 20
• frame_performance_summary

  Overall frame health metrics

  Try:Summarize frame performance and report jank rate and P99 CPU time
• thread_contention_analyzer

  Find synchronization bottlenecks

  Try:Find lock contention between 15s–30s and show worst waits
• binder_transaction_profiler

  Analyze Binder IPC performance

  Try:Profile slow Binder transactions and group by server process
• memory_leak_detector

  Find sustained memory growth patterns

  Try:Detect memory-leak signals over the last 60s
• heap_dominator_tree_analyzer

  Identify memory-hogging classes

  Try:Analyze heap dominator classes and list top offenders

Comparable tools

Installation

{
  "mcpServers": {
    "perfetto-mcp": {
      "command": "uvx",
      "args": ["perfetto-mcp"]
    }
  }
}

claude mcp add perfetto-mcp --scope user -- uvx perfetto-mcp

{
  "mcpServers": {
    "perfetto-mcp": {
      "command": "uvx",
      "args": ["perfetto-mcp"]
    }
  }
}

pip3 install perfetto-mcp
python3 -m perfetto_mcp

FAQ

What file format does Perfetto MCP work with?

Perfetto MCP works with Perfetto trace files (.perfetto-trace format).

Can I use a local trace_processor_shell binary?

Yes, set the PERFETTO_MCP_TRACE_PROCESSOR_BIN_PATH environment variable to the path of your local binary.

What types of performance analyses can I perform?

You can analyze ANRs, CPU usage, frame jank, memory leaks, thread contention, binder transactions, and more using natural language prompts.

Compare perfetto-mcp with