test-quality-analysis
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and analyzes external code files.\n
- Ingestion points: The skill analyzes code using tools such as
Read,Grep, andGlobto detect patterns and smells within files.\n - Boundary markers: The instructions do not specify any boundary markers to isolate the untrusted code being analyzed from the agent's control logic.\n
- Capability inventory: The agent has access to
Bash,Write, andEdittools, allowing for file modification and command execution if instructions in analyzed data are followed.\n - Sanitization: There is no evidence of content sanitization or validation for the code files processed by the skill.
Audit Metadata