The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data from Instagram, which is an untrusted source that could contain malicious instructions. * Ingestion points: The agent retrieves data from Instagram using the Media, Comments, and Replies tables. * Boundary markers: There are no explicit delimiters or instructions provided to the agent to differentiate between system instructions and data fetched from external tables. * Capability inventory: The skill allows the agent to perform SQL SELECT operations to read data and INSERT operations to post comments or replies, creating a feedback loop. * Sanitization: The skill does not specify any sanitization, filtering, or validation for the content retrieved from Instagram.

connect-ai-instagram