profile-sculptor-backend

Capture thread dumps, viztracer/Perfetto traces, and CPU samples from Sculptor backends.

213|13|Updated Aug 7, 2025
One-click install
npx skills add https://github.com/imbue-ai/sculptor --skill profile-sculptor-backend
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: profile-sculptor-backend
Source: https://github.com/imbue-ai/sculptor/tree/main/.claude/skills/profile-sculptor-backend
Command: npx skills add https://github.com/imbue-ai/sculptor --skill profile-sculptor-backend

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Quickly analyze and profile the performance of a running Sculptor backend with instant thread dumps and detailed traces without additional privileges.

Core Features & Use Cases

  • Instant Thread Dumps: Provides instant thread dumps for diagnosing issues within a Sculptor backend.
  • Visual Tracing and Traces: Generates viztracer/Perfetto traces for in-depth analysis.
  • CPU Sampling: Offers CPU sampling for creating flamegraphs and native stacks.
  • Use Case: When encountering issues with a Sculptor backend, such as a deadlock or high CPU usage, this skill allows you to gain insight into the process to diagnose and fix problems efficiently.

Quick Start

To profile the Sculptor backend with instant thread dumps, run the command 'sculpt debug threads'.

Frequently Asked Questions about profile-sculptor-backend

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get a thread dump from a running backend to diagnose a deadlock?

To perform CPU profiling on a backend, this skill offers CPU sampling capabilities to capture native stacks and generate data for flamegraphs, helping you identify and resolve CPU-intensive processes efficiently.

Can I generate Perfetto traces for backend analysis without extra privileges?

Yes, you can generate viztracer and Perfetto traces for in-depth backend analysis directly without needing additional privileges, allowing you to visualize performance issues within the Sculptor environment.

What is the best way to profile high CPU usage in a Python backend?

The best way to profile high CPU usage is using CPU sampling to create flamegraphs and native stacks, providing detailed Python-level diagnostics to pinpoint and fix CPU-intensive processes.

Do I need special permissions to run diagnostics on a Sculptor backend?

No, you do not need special permissions to run diagnostics on a Sculptor backend, as the skill is optimized to capture thread dumps and traces without requiring additional privileges.