The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by instructing the agent to automatically ingest data from external context files to inform its behavior.
Ingestion points: In SKILL.md, the instructions mandate reading .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md before proceeding with the task.
Boundary markers: The skill lacks explicit boundary markers or "ignore embedded instructions" warnings for the agent when processing these external files.
Capability inventory: The agent has the capability to generate complex business logic, including lifecycle stages, lead scoring models, and automation workflows, based on the content of these files.
Sanitization: No sanitization, validation, or escaping logic is defined for the content ingested from the context files.

revops