create-prd
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill instructions and scripts do not exhibit malicious behavior or security policy violations.
- [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for processing untrusted user input during the PRD creation process. Ingestion points: User-provided product name, objective, and market segments gathered in the 'Clarify First' section of SKILL.md. Boundary markers: Not utilized; the inputs are directly interpolated into the markdown output and script arguments. Capability inventory: Local file writing via the prd_scaffolder.py script and data handoff to downstream project management skills. Sanitization: No explicit content filtering or validation of the input strings is implemented.
- [COMMAND_EXECUTION]: The skill uses a provided Python script to generate document skeletons. The script 'scripts/prd_scaffolder.py' relies solely on Python standard libraries (argparse, sys, textwrap, datetime). Execution is limited to local scaffolding and optional file output as documented in the skill instructions.
Audit Metadata