Model Resource Profiler

PassAudited by ClawScan on May 1, 2026.

Overview

The skill appears to be a purpose-aligned local profiler analyzer that reads user-provided JSON/JSON.GZ files and writes reports, with no evidence of credential use, network exfiltration, persistence, or unrelated system access.

This skill looks appropriate for local analysis of trusted PyTorch profiler exports. Before installing or using it, confirm you are comfortable with it running the included Python script on the profiler files you provide, and avoid giving it unrelated private files.

Findings (1)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

The agent may execute the included analyzer script to read selected profiler artifacts and write markdown or JSON reports.

Why it was flagged

The workflow explicitly runs a bundled local Python script. This is expected for this analysis skill and is bounded to user-provided JSON/JSON.GZ artifacts, but it is still local code execution the user should recognize.

Skill content
Run deterministic analysis script. - Use `scripts/analyze_profile.py` for summary extraction.
Recommendation

Run it only on trusted JSON/JSON.GZ profiler files, choose output paths you recognize, and review the generated report before applying optimization changes.