The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided descriptions and read from a local context file (.claude/product-marketing-context.md), which establishes a surface for potential indirect prompt injection. * Ingestion points: The agent is instructed to read .claude/product-marketing-context.md and gather product/audience details from user queries. * Boundary markers: No delimiters or specific instructions to isolate or ignore instructions within the ingested data are present. * Capability inventory: The skill's primary function is text generation; it refers to other skills but lacks direct system execution or network capabilities. * Sanitization: There are no explicit instructions for validating or sanitizing the information gathered from external sources.

copywriting