customer-persona
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
infshCLI tool via theBashtool to perform various tasks including target market research and image generation through specified applications liketavily/search-assistantandfalai/flux-dev-lora.- [EXTERNAL_DOWNLOADS]: The skill references an external markdown file on the vendor's GitHub repository (inference-sh/skills) containing installation instructions for their CLI tool.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and processes data from external search engines. - Ingestion points: External content is retrieved via
tavily/search-assistant,exa/search, andexa/answerapps as shown inSKILL.md. - Boundary markers: No explicit delimiter or "ignore instructions" markers are used in the prompt templates.
- Capability inventory: The skill has the capability to execute system commands via the
Bashtool (specifically theinfshCLI). - Sanitization: There is no evidence of output sanitization or filtering for the data returned by the search apps.
Audit Metadata