pipefy-html-interfaces
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes HTML documents from the Pipefy platform, which could contain malicious instructions intended to influence the agent.\n
- Ingestion points: The
get_interface_htmltool (SKILL.md) reads existing HTML documents into the agent's context.\n - Boundary markers: The instructions do not define specific delimiters or warnings to ignore embedded instructions within the HTML content.\n
- Capability inventory: The skill can create, modify, and delete interfaces using
create_html_interface,write_interface_html, anddelete_portal.\n - Sanitization: The skill mentions a platform-side validator for HTML attributes but lacks explicit sanitization for natural language instructions in the HTML body or comments.\n- [COMMAND_EXECUTION]: The skill uses the
pipefyCLI tool for organization resource management. This is a vendor-provided tool used for its intended purpose.
Audit Metadata