business-requirements-interview

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the sherpy command-line utility to perform strict validation of the generated requirements document at the end of the interview process.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes and stores untrusted user data.
  • Ingestion points: Project descriptions provided by the user and subsequent interview answers, as well as the business-interview.jsonl file when resuming a session.
  • Boundary markers: None present; the skill lacks specific delimiters or instructions to ignore commands embedded within user-supplied project details.
  • Capability inventory: The skill writes to the file system (artifacts and requirements directories) and executes local commands via the sherpy validator.
  • Sanitization: No sanitization or validation of the content of user answers is performed before they are interpolated into the final YAML output or the session log.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:29 PM
Security Audit — agent-trust-hub — business-requirements-interview