perfetto-trace-analysis

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the trace_processor utility from https://get.perfetto.dev/trace_processor. This source is a well-known service associated with the official Perfetto project.
  • [REMOTE_CODE_EXECUTION]: The skill downloads and then executes a Python wrapper script (trace_processor). This tool is essential for the skill's primary function of trace analysis and originates from a well-known service.
  • [COMMAND_EXECUTION]: The skill invokes shell commands such as curl, chmod, and the ./trace_processor script to perform analysis tasks. These operations are limited to the workspace and are standard for debugging tools.
  • [PROMPT_INJECTION]: The skill permits the execution of user-provided SQL queries, which is a potential surface for indirect prompt injection. This is mitigated by a mandatory validation loop.
  • Ingestion points: External Perfetto trace files and user-provided SQL queries.
  • Boundary markers: Not explicitly present for trace data; uses a step-by-step verification process for SQL queries.
  • Capability inventory: Subprocess execution of the trace_processor analysis tool.
  • Sanitization: Includes an 'Execution Protocol' with a checklist (Step 2) to validate SQLite syntax, table existence, and safety constraints before any query is run.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 03:18 PM
Security Audit — agent-trust-hub — perfetto-trace-analysis