The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a potential for indirect prompt injection by ingesting untrusted data from the 'active project' or 'active persona' to construct research queries.
Ingestion points: The skill dynamically pulls 'genre', 'subgenre', and 'target audience' metadata from the user's project context (SKILL.md).
Boundary markers: There are no explicit delimiters or 'ignore' instructions mentioned to prevent malicious content in project files from overriding the research agent's behavior.
Capability inventory: The skill utilizes file_read and file_write permissions to access project data and save reports to the workspace. It also invokes the research-lookup tool for live-web access.
Sanitization: The instructions do not define any validation or filtering logic for the data interpolated into the search query templates.

marketing-research