perfetto-sql

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the trace_processor wrapper script from the official Perfetto domain (https://get.perfetto.dev/trace_processor), which is a well-known service maintained by Google.
  • [COMMAND_EXECUTION]: The execution protocol involves using curl to fetch the tool, chmod +x to modify permissions, and executing the script via the system shell to process trace data.
  • [REMOTE_CODE_EXECUTION]: The downloaded trace_processor script is a Python wrapper that lazy-loads precompiled binaries from Google's infrastructure upon execution.
  • [PROMPT_INJECTION]: The skill instructs the agent to use user-provided SQL queries without modification in Step 1. Ingestion point: User-provided SQL via the Precedence Rule; Boundary markers: Absent; Capability inventory: Shell execution of the trace_processor CLI; Sanitization: Absent for direct user queries. This creates a surface for indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 03:18 PM
Security Audit — agent-trust-hub — perfetto-sql