The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest data from an external context file, creating an indirect prompt injection vulnerability surface if that file contains untrusted content. \n
Ingestion points: The skill instructions in SKILL.md direct the agent to read .agents/product-marketing-context.md or .claude/product-marketing-context.md before proceeding. \n
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the content of the marketing context file as untrusted or to ignore any embedded directives. \n
Capability inventory: The skill is restricted to providing SEO guidance and generating content templates; no executable code, filesystem modification, or network-active tools are included in the provided scripts. \n
Sanitization: No sanitization, validation, or filtering of the external file's content is implemented within the skill logic.

ai-seo