The Agent Skills Directory

[PROMPT_INJECTION]: The skill utilizes the web_fetch tool to ingest content from Amazon product pages (BSR, reviews, ratings) to perform market analysis. This ingestion of untrusted external data presents a surface for indirect prompt injection, where an attacker could place malicious instructions in product metadata or reviews to influence the agent's logic.
Ingestion points: Amazon product pages accessed via web_fetch and web_search.
Boundary markers: Absent; there are no specific delimiters or safety instructions defined to isolate the fetched web content from the agent's primary instructions.
Capability inventory: The skill uses web_search and web_fetch to retrieve data and generates structured analysis reports.
Sanitization: There are no explicit instructions for the agent to sanitize or validate the integrity of the fetched web data before processing it.

amazon-niche-finder