The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external HTML code and rendered components for auditing purposes, creating a potential vector for instructions embedded in data. * Ingestion points: The instructions in SKILL.md and references/rule.md involve reviewing "rendered markup" and "code examples." * Boundary markers: There are no specific instructions or delimiters provided to ensure the agent ignores instructions that may be contained within the input HTML. * Capability inventory: The skill's scope is restricted to descriptive analysis, flagging violations, and providing explanations; it does not request tools for network access, file system modifications, or command execution. * Sanitization: No methods for sanitizing or escaping the analyzed HTML are mentioned in the instructions.

list-structure