What outputs does perf-profiler generate?

A structured report with tool, command, hotspots (file:line), and artifacts (flame graphs or profiles).

Why verify debug symbols?

Ensures accurate symbol names and reliable hotspots mapping in your reports.

Can perf-profiler work with non-native apps?

Yes, as long as you can run a profiler and capture hotspot data; follow docs/perf-requirements.md for symbol and scenario requirements.

perf-profiler

Scanned

npx machina-cli add skill ComposioHQ/awesome-claude-plugins/profile --openclaw

Files (1)

SKILL.md

726 B

perf-profiler

Run profiling tools and capture hotspots with evidence.

Follow docs/perf-requirements.md as the canonical contract.

Required Rules

Verify debug symbols before profiling.
Capture file:line for hotspots.
Provide flame graph or equivalent output when possible.

Output Format

tool: <profiler>
command: <command>
hotspots:
  - file:line - reason
artifacts:
  - <path to flame graph or profile>

Constraints

No profiling without a clear scenario.
Keep outputs minimal and evidence-backed.

Source

git clone https://github.com/ComposioHQ/awesome-claude-plugins/blob/master/perf/skills/profile/SKILL.mdView on GitHub

Overview

perf-profiler runs profiling tools to identify CPU and memory hot paths, producing flame graphs and evidence-ready outputs. It enforces perf requirements and emphasizes symbol accuracy and precise hotspots.

How This Skill Works

The tool executes a profiler on the target command, verifies debug symbols before profiling, and collects hotspots with file:line information. It then outputs a minimal, evidence-backed report containing the profiler used, the command executed, hotspot entries, and paths to artifacts such as flame graphs.

When to Use It

Investigating CPU hotspots in a critical request path
Tracking memory allocation hotspots during peak load
Generating flame graphs to visualize bottlenecks
Collecting JFR or perf evidence for JVM-native integrations
Ensuring evidence-backed profiling with verified symbols

Quick Start

Step 1: Run perf-profiler with a target tool and command
Step 2: Verify debug symbols are present and profiling is configured
Step 3: Review hotspots and artifacts (flame graphs) from the output

Best Practices

Verify debug symbols before profiling
Capture file:line granularity for all hotspots
Have a clear profiling scenario and goals
Provide a flame graph or equivalent output when possible
Keep outputs minimal and evidence-backed for stakeholders

Example Use Cases

Profile a web API latency spike to locate hot request paths
Profile a data-processing job to find heavy memory allocations
Profile a microservice under load to understand bottlenecks
Capture JFR data for a JVM service that interfaces with native code
Generate a flame graph to present to the performance team

Frequently Asked Questions

Add this skill to your agents