The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/lighthouse_audit.py executes the lighthouse CLI tool to perform audits. It utilizes subprocess.run with a list of arguments rather than a single string, which is a secure practice that prevents shell injection attacks.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it ingests external data (from a target URL) and processes the results for the agent.
Ingestion points: The url command-line argument processed in scripts/lighthouse_audit.py.
Boundary markers: The skill does not use specific delimiters to separate external audit data from agent instructions.
Capability inventory: The skill has the ability to execute the lighthouse command via subprocess.run (scripts/lighthouse_audit.py).
Sanitization: There is no validation or sanitization of the input URL or the resulting JSON report before the data is processed and presented to the agent.

performance-profiling