NYC

performance-profiling

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • COMMAND_EXECUTION (SAFE): The scripts/lighthouse_audit.py script executes the lighthouse CLI using a safe argument-list approach in subprocess.run, effectively preventing shell injection vulnerabilities.
  • Indirect Prompt Injection (SAFE): The skill audits content from external URLs. It mitigates injection risks by parsing the tool's JSON output into a strict local schema and returning only numeric scores and a script-generated summary to the agent, rather than the raw website content.
  • Ingestion points: scripts/lighthouse_audit.py (via external URL fetching)
  • Boundary markers: Not applicable due to strict output filtering
  • Capability inventory: Bash (allowed-tool), subprocess.run
  • Sanitization: High; only score integers and a pre-defined status message are extracted and returned.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:44 PM