linkedin-content
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation instructions from the official GitHub repository of
inference-sh. - [COMMAND_EXECUTION]: The skill utilizes the
beltCLI tool for legitimate operations such as authentication, content research via search assistants, and cross-platform posting. The execution environment is restricted to thebeltcommand via the skill frontmatter. - [INDIRECT_PROMPT_INJECTION]: The skill includes an ingestion surface for untrusted data by running search queries through a search assistant. However, this is used for standard content research, and the skill's capabilities are focused on text and image generation rather than sensitive system operations.
Audit Metadata