customer-persona
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install external CLI tools and platform-specific skills.
- The instructions include installing the
belt-sh/clivianpxand referencing an external installation guide athttps://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md. - Related skills from the
inference-shrepository are also recommended for installation. - [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute the
beltCLI for running remote market research and image generation applications. - Commands like
belt app runare used to interface with external services including Tavily, Exa, and Fal.ai. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external search providers, creating a potential vector for indirect instructions to influence the agent.
- Ingestion points: External data enters the context through
tavily/search-assistant,exa/search, andexa/answerapp outputs as described inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the search results are provided.
- Capability inventory: The skill possesses the capability to execute shell commands via the Bash tool (
beltcommand) as specified in theallowed-toolsmetadata. - Sanitization: There is no evidence of sanitization or validation of the content retrieved from external search queries before it is processed to create the persona.
Audit Metadata