The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface by ingesting untrusted data from the web and writing it to local storage. * Ingestion points: The skill uses WebSearch and WebFetch tools to retrieve external data. * Boundary markers: There are no instructions to use delimiters or ignore embedded instructions within the fetched content. * Capability inventory: The skill specifies writing the output to the project's 02-research/ folder. * Sanitization: The instructions do not mandate escaping, validation, or filtering of the external content before it is processed and saved.

market-research