The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill analyzes user-provided source code to identify performance issues, which constitutes an ingestion point for potentially untrusted data. If the code being reviewed contains adversarial instructions hidden in comments or strings, the agent might inadvertently execute them.
Ingestion points: Project code files and snippets provided by the user for performance review as described in SKILL.md.
Boundary markers: The skill does not define specific delimiters or instructions for the agent to use to isolate the user-provided code from the agent's internal instruction context.
Capability inventory: The skill is intended for use with an agent that has capabilities to modify the filesystem and interact with databases.
Sanitization: No sanitization or escaping mechanisms are specified to filter out potential prompt injection patterns within the code under review.

performance-optimizer