trace-sql

Analyze Pulp Perfetto .pftrace files with SQL to identify performance bottlenecks.

13|1|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/Generous-Corp/pulp --skill trace-sql-generous-corp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trace-sql
Source: https://github.com/Generous-Corp/pulp/tree/main/.agents/skills/trace-sql
Command: npx skills add https://github.com/Generous-Corp/pulp --skill trace-sql-generous-corp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill resolves performance bottlenecks in Pulp applications by providing a standardized SQL-based methodology for querying and analyzing Perfetto trace files.

Core Features & Use Cases

  • Standardized Querying: Provides a robust SQL discipline for analyzing .pftrace files, including idempotent view definitions and stable key joins.
  • Performance Primitives: Ships with a trace-stdlib containing pre-authored views for identifying slowest frames, DSP node costs, frame budget overruns, and xruns.
  • Use Case: When a user reports audio glitches or UI hitches, use this skill to query the trace for frame budget overruns or DSP node hotspots to pinpoint the exact cause of the performance degradation.

Quick Start

Use the trace-sql skill to query the provided trace file for the slowest frames and identify potential performance bottlenecks.

Frequently Asked Questions about trace-sql

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

FAQPage Schema
How do I analyze Perfetto trace files to find audio glitches and UI hitches?

SQL trace analysis identifies performance bottlenecks in Pulp .pftrace files by querying for slowest frames, DSP node costs, and frame budget overruns. It applies standardized SQL queries to pinpoint exact causes of audio glitches and UI rendering hitches.

What is the best way to debug frame-pipeline latency in Pulp applications?

The best way to debug frame-pipeline latency is applying SQL queries to Perfetto trace captures. This skill queries for xruns and frame budget overruns to isolate resource usage patterns and specific DSP node costs causing the latency.

Can I use SQL to query .pftrace files for specific performance metrics?

Yes, you can query .pftrace files using SQL through trace_processor integration. This skill executes idempotent SQL queries against Perfetto captures to extract metrics like slowest frames, DSP node costs, frame budget overruns, and xruns.

Does analyzing Pulp performance bottlenecks require any specific environment setup?

Analyzing Pulp performance bottlenecks requires trace_processor integration to execute SQL queries against .pftrace files. You must provide Perfetto trace captures from your Pulp application to query slowest frames and resource usage patterns.

How do I identify which DSP nodes are causing performance degradation in my Pulp app?

To identify DSP nodes causing performance degradation, query your Perfetto trace for DSP node costs and frame budget overruns. This skill provides pre-authored trace-stdlib views to isolate these hotspots within the Pulp framework.

Why do I need idempotent view definitions when analyzing Perfetto traces?

Idempotent view definitions ensure SQL trace analysis returns consistent, repeatable results when querying .pftrace files. This standardized SQL discipline maintains stable key joins, preventing inconsistent data extraction during performance debugging.