The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions recommend installing the baneh-mcp package using the uvx tool. This package, provided by the author, contains the logic for interacting with the Baneh website's APIs.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from an external website, specifically product descriptions and buyer reviews, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Product data, descriptions, and reviews are fetched from external APIs via tools like bn_product and bn_reviews.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within the fetched content.
Capability inventory: The skill is primarily read-only for information retrieval and does not have capabilities for file modification or administrative shell access.
Sanitization: No content sanitization or validation routines are specified for the data retrieved from the website.

baneh