hermoso-ad-from-brand
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npx -y hermosoto download and run the vendor's CLI tool directly from the npm registry during execution. - [COMMAND_EXECUTION]: The skill requires the
Bashtool to execute shell commands for authenticating with the service, onboarding brands, and generating ad creative. - [DYNAMIC_EXECUTION]: The CLI includes functionality (
hermoso call <name>) to dynamically discover and execute tools within the Hermoso ecosystem based on search results or specific needs at runtime. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites (
--domain) and social media platforms (--social) to create brand profiles and ad concepts. - Ingestion points: Data enters the context via
hermoso brand draftthrough domain scraping and social media handles defined inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded content are present in the command strings.
- Capability inventory: The
Bashtool is used for file system operations (fetching media viahermoso fetch) and network communication via the CLI. - Sanitization: The instructions do not specify manual sanitization, relying on the vendor CLI's internal processing of external data.
Audit Metadata