The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing user-supplied plans, decisions, or ideas which are untrusted data sources. * Ingestion points: User-provided input defining the plan or idea to be 'grilled' as described in SKILL.md. * Boundary markers: The skill does not define specific delimiters or warnings to prevent the agent from following instructions embedded within the user's idea. * Capability inventory: The skill directs the agent to use sub-agents and environment tools to access the filesystem for fact-finding. * Sanitization: There are no instructions for sanitizing or validating user input before it is processed by the agent or its tools.

grilling