The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data sources, including Meta Ads Manager exports and Events Manager screenshots. This creates a potential surface for indirect prompt injection, where malicious instructions could be embedded in advertising campaign data to influence the agent's behavior. * Ingestion points: Meta Ads Manager exports, Events Manager screenshots, and EMQ (Event Match Quality) scores. * Boundary markers: The skill does not define specific delimiters or instructions to treat the ingested data as untrusted content. * Capability inventory: The skill's primary capabilities are reading local reference files, performing calculations, and writing a report (META-ADS-REPORT.md). No network or system-level execution capabilities are present in the provided instructions. * Sanitization: No sanitization or validation logic is provided for the external advertising data before it is processed by the agent.

ads-meta