The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, specifically competitor reviews (1-3 star ratings) and user-supplied product information, which creates a surface for indirect prompt injection attacks.
Ingestion points: The skill processes data from products.md, learnings.jsonl, and external evidence sources like customer reviews, YouTube comments, and interview transcripts.
Boundary markers: The instructions do not specify the use of delimiters (e.g., XML tags or triple quotes) or explicit "ignore embedded instructions" warnings when processing this untrusted content.
Capability inventory: Across its scripts, the skill has the capability to write data back to the local learnings.jsonl file to persist "memory."
Sanitization: There is no mention of sanitizing, filtering, or escaping external inputs before they are interpolated into the agent's context for analysis.

afa-product