The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted requirement descriptions and using them to drive file-writing capabilities.
Ingestion points: The skill ingests data from the user's initial requirement description (Step 1) and subsequent interactive responses (Step 3) as described in the clarification process within SKILL.md.
Boundary markers: The instructions lack explicit boundary markers or delimiters to differentiate between user-provided data and agent instructions, which could allow malicious input to influence the agent's behavior during the document generation phase.
Capability inventory: The skill utilizes the Write tool to generate and save PRD files to the ./docs/prds/ directory, providing a mechanism for an injection to result in the creation of files with untrusted content.
Sanitization: There is no mention of input validation, sanitization, or safety filtering applied to the user-supplied content before it is interpolated into the final PRD output.

requirements-clarity