vertical-creator
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided vertical definition packages, including JSON configurations and Markdown templates, for creation and review tasks. This represents an indirect prompt injection surface where malicious instructions could be embedded in the data provided to the agent.
- Ingestion points: User-supplied vertical directories and individual asset files like vertical.json and template Markdown files.
- Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore natural language instructions found within the vertical assets it reviews.
- Capability inventory: The skill employs shell command execution through the ha vertical validate tool and performs filesystem operations to manage vertical assets.
- Sanitization: The process relies on schema validation via the ha CLI, which mitigates structural issues but does not necessarily filter prompt injection attempts within template content.
Audit Metadata