skills/jeremylongshore/claude-code-plugins-plus-skills/providing-performance-optimization-advice/Gen Agent Trust Hub
providing-performance-optimization-advice
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- Metadata Poisoning (MEDIUM): The skill claims to provide performance profiling and optimization impact estimation, but the provided Python scripts are identical templates that only perform basic file size and JSON validation. This is deceptive as it does not fulfill the stated purpose of the skill.\n- Indirect Prompt Injection (LOW): The skill's primary function is to analyze external codebases, which exposes the agent to malicious instructions embedded in data.\n
- Ingestion points: Codebase files in
{baseDir}/accessed via scripts andReadtools.\n - Boundary markers: Absent. No instructions exist to prevent the agent from executing instructions found within the code being analyzed.\n
- Capability inventory: Access to
Bash,Write, andEdittools provides a high-impact exploitation surface.\n - Sanitization: Absent. There is no logic to filter or sanitize input from the target codebase.
Audit Metadata