p-image
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation documentation for the infsh CLI from the vendor's GitHub repository (inference-sh/skills).\n- [PROMPT_INJECTION]: The skill processes user-supplied text prompts which presents a surface for indirect prompt injection.\n
- Ingestion points: Untrusted data enters via the 'prompt' field in the JSON input within SKILL.md.\n
- Boundary markers: Data is structured in JSON format which provides limited delimitation.\n
- Capability inventory: The skill utilizes the 'infsh' CLI tool (Bash) to execute commands that communicate with external APIs.\n
- Sanitization: No input validation or sanitization of the prompt is described in the skill instructions.
Audit Metadata