The Agent Skills Directory

[PROMPT_INJECTION]: The skill acts as an ingestion point for untrusted user input during the requirement clarification interview. This data is processed and saved to docs/product/.ompm/clarification-result.json, which is then utilized by downstream agent tools like prd-gen. This workflow presents an indirect prompt injection surface where a user could provide instructions disguised as requirements to manipulate the behavior of subsequent tools.
Ingestion points: All user requirement descriptions and interview responses defined in SKILL.md and processed at runtime.
Boundary markers: The skill uses status markers (✅ for confirmed, ⚠️ for pending, ❓ for unverified) to track data provenance, but these markers are intended for process flow and do not serve as security boundaries or sanitization filters.
Capability inventory: The skill reads from and writes to the docs/product/ directory to manage product documentation.
Sanitization: There is no evidence of input sanitization, filtering, or escaping of user-provided content before it is stored or passed to the next stage in the pipeline.

clarify-requirements