optimize
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source code to provide performance optimizations, which creates an opportunity for indirect prompt injection if the code contains malicious instructions.\n- Ingestion points: User-provided code files and the
[target]argument (SKILL.md).\n- Boundary markers: Absent. No clear separation between untrusted code and agent instructions is defined.\n- Capability inventory: The agent uses file-writing and execution capabilities to implement changes.\n- Sanitization: Absent. Input code is not validated for embedded malicious prompts.\n- [NO_CODE]: No executable scripts or binaries are distributed with the skill; it consists entirely of instructional markdown.
Audit Metadata