The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a Python script scripts/lighthouse_audit.py that executes the lighthouse command-line tool using subprocess.run. The command is constructed as a list of arguments, which is a secure practice to prevent shell injection, though it lacks explicit validation of the URL parameter beyond passing it as a positional argument.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data by reading JSON reports generated by the Lighthouse CLI. While this represents an attack surface for indirect prompt injection (where malicious content in a web page might theoretically influence the audit report content), the script only extracts specific numeric scores and generates a summary based on predefined logic, significantly limiting any potential impact.
Ingestion points: Reads temporary JSON files generated by the lighthouse CLI in scripts/lighthouse_audit.py.
Boundary markers: None explicitly present in the data ingestion phase.
Capability inventory: Uses subprocess.run to call the auditor and os.unlink for temporary file cleanup.
Sanitization: The script performs type conversion (casting scores to integers) and uses a hardcoded conditional structure for summary generation, which mitigates simple text-based injection.

performance-profiling