The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes untrusted data from requirement_template.md and external .xmind files to generate PRD content. Malicious instructions embedded in these files could attempt to manipulate the agent's behavior during the extraction or generation phase.
[COMMAND_EXECUTION]: The helper script scripts/utils.js provides file system capabilities (readFile, writeFile, mkdir) using the Node.js fs module. The lack of path sanitization when deriving filenames from project names (extracted from untrusted templates) creates a potential path traversal surface, though the mandatory human review process defined in templates/review_process.md serves as a significant security control.

prd-suite