The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to take natural language descriptions from the user and project metadata to determine which high-capability skill or workflow to execute next.
Ingestion points: The skill ingests raw user input (natural language descriptions of PM tasks) and reads state information from docs/product/.ompm/current-workflow.json and other files within the docs/product/ directory.
Boundary markers: The instructions lack explicit boundary markers or directives for downstream skills to ignore potentially malicious instructions embedded within the user-provided descriptions.
Capability inventory: The skill has the authority to trigger various workflows with high capabilities, including file system writes (prd-gen), interactive prototyping (prototype-design), and market research (market-research).
Sanitization: There is no evidence of input sanitization, filtering, or validation before the user's input is processed for routing or passed to the next skill in the workflow.

ompm