The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project context files like .agents/product-marketing.md and live web research results to inform its persona simulations. This external data could potentially contain malicious instructions designed to manipulate the agent's behavior.
Ingestion points: The skill reads .agents/product-marketing.md for context and performs web searches to update advisor stances.
Boundary markers: There are no specific instructions or delimiters used to separate the external context from the skill's core system instructions.
Capability inventory: The skill utilizes file reading/writing (within the project scope for advisor dossiers) and network operations (via research tools).
Sanitization: No technical sanitization or validation of the ingested marketing data or search results is defined.

marketing-council