How do I install the Torchprofiler Trace Analysis skill?

Run npx skillmds@latest add shan2l/torchprofiler-trace-analysis in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Torchprofiler Trace Analysis safe to use?

SkillMD's automated safety review verdict for this skill is PASS. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: executes scripts, makes network calls, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Torchprofiler Trace Analysis?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Torchprofiler Trace Analysis free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Torchprofiler Trace Analysis?

Shan2L (@shan2l) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Torchprofiler Trace Analysis by shan2l

Name: Torchprofiler Trace Analysis
Rating: 4.5 (4 reviews)
Author: Shan2L

Torchprofiler Trace Analysis

Analyze PyTorch profiler and vLLM trace files with Perfetto SQL evidence. Use when working with torch profiler traces, merged_trace files, vLLM profiling logs, Perfetto trace analysis, GPU kernel hotspots, synchronization waits, rank imbalance, prefill/decode bottlenecks, or optimization reports.

Shan2L Updated Aug 19, 2026 0 repo stars

File contents

TorchProfiler Trace Analysis

Purpose

Use this skill to analyze PyTorch profiler traces from vLLM/VAP runs. Prefer evidence from Perfetto SQL and structured trace metadata over raw JSON snippets.

Attribution

This project is inspired by the evidence-driven workflow design of:

Gracker/Perfetto-Skills: standard Agent Skill structure, workflow routing, SQL-backed evidence, and report contracts.
Gracker/SmartPerfetto: AI-assisted Perfetto analysis, evidence workflows, reports, and trace-processor-backed SQL analysis.

No code or SQL is copied from those projects. The SQL presets and workflows in this project are original and specialized for PyTorch profiler / vLLM traces.

Workflow

Identify the latest or requested trace.
Prefer merged traces named like *-merged_trace.json or *-merged_trace.json.gz.
Run trace overview queries first.
Collect evidence for:
- synchronization waits
- GPU kernel hotspots
- CPU operator hotspots
- rank imbalance
- memory copies
- prefill/decode spans
- idle gaps
Summarize findings with evidence.
Separate confirmed evidence from hypotheses.
Recommend next inspections in Perfetto/TensorBoard.

Report Format

Use this structure:

# TorchProfiler Trace Report

## Executive Summary
Short summary of the most likely bottleneck.

## Trace Metadata
- Trace file:
- Merged trace:
- Event count:
- Time span:
- Ranks:

## Evidence
| Area | Evidence | Interpretation |
|---|---|---|

## Bottleneck Hypotheses
1. Hypothesis with supporting evidence.

## Perfetto Inspection Guide
- Tracks/events to inspect next.

## Optimization Suggestions
- Concrete model/config/benchmark changes to try.

## Evidence Gaps
- Missing data or uncertainty.

Safety

Do not infer causal conclusions from a query that only parsed successfully. Label uncertain conclusions as hypotheses. Do not request raw trace JSON unless a focused preview is necessary.

Utilities

Query presets: queries.yaml
Runner: runner.py
Workflow details: WORKFLOWS.md

---name: torchprofiler-trace-analysisdescription: Analyze PyTorch profiler and vLLM trace files with Perfetto SQL evidence. Use when working with torch profiler traces, merged_trace files, vLLM profiling logs, Perfetto trace analysis, GPU kernel hotspots, synchronization waits, rank imbalance, prefill/decode bottlenecks, or optimization reports.---# TorchProfiler Trace Analysis## PurposeUse this skill to analyze PyTorch profiler traces from vLLM/VAP runs. Prefer evidence from Perfetto SQL and structured trace metadata over raw JSON snippets.## AttributionThis project is inspired by the evidence-driven workflow design of:- `Gracker/Perfetto-Skills`: standard Agent Skill structure, workflow routing, SQL-backed evidence, and report contracts.- `Gracker/SmartPerfetto`: AI-assisted Perfetto analysis, evidence workflows, reports, and trace-processor-backed SQL analysis.No code or SQL is copied from those projects. The SQL presets and workflows in this project are original and specialized for PyTorch profiler / vLLM traces.## Workflow1. Identify the latest or requested trace.2. Prefer merged traces named like `*-merged_trace.json` or `*-merged_trace.json.gz`.3. Run trace overview queries first.4. Collect evidence for: - synchronization waits - GPU kernel hotspots - CPU operator hotspots - rank imbalance - memory copies - prefill/decode spans - idle gaps5. Summarize findings with evidence.6. Separate confirmed evidence from hypotheses.7. Recommend next inspections in Perfetto/TensorBoard.## Report FormatUse this structure:```markdown# TorchProfiler Trace Report## Executive SummaryShort summary of the most likely bottleneck.## Trace Metadata- Trace file:- Merged trace:- Event count:- Time span:- Ranks:## Evidence| Area | Evidence | Interpretation ||---|---|---|## Bottleneck Hypotheses1. Hypothesis with supporting evidence.## Perfetto Inspection Guide- Tracks/events to inspect next.## Optimization Suggestions- Concrete model/config/benchmark changes to try.## Evidence Gaps- Missing data or uncertainty.```## SafetyDo not infer causal conclusions from a query that only parsed successfully. Label uncertain conclusions as hypotheses. Do not request raw trace JSON unless a focused preview is necessary.## Utilities- Query presets: [queries.yaml](queries.yaml)- Runner: [runner.py](runner.py)- Workflow details: [WORKFLOWS.md](WORKFLOWS.md)

Frequently asked questions