fresh-eyes
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions and markdown templates for organizing thoughts during troubleshooting. It does not initiate any system changes or network connections.
- [NO_CODE]: Analysis of the skill file confirms the absence of any scripts (Python, JavaScript, Shell), binary executables, or command-line instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process conversation history, which is technically untrusted data. However, as the skill possesses no functional capabilities or tools, this does not present an exploitable risk.\n
- Ingestion points: Reads and summarizes previous conversation turns and user-provided debugging attempts in SKILL.md.\n
- Boundary markers: Employs structured markdown headers and tables to organize the 'Fresh Eyes' reset output.\n
- Capability inventory: None. The skill does not have access to tools, shell execution, file system writes, or network APIs.\n
- Sanitization: No specific input sanitization is performed, as the output is restricted to standard markdown text within the agent response.
Audit Metadata