The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local Python proxy client (scripts/amazon_search.py) to interact with the Amazon search API. The instructions guide the agent to execute this script via the command line with specific JSON parameters to retrieve and process product data.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted product information from external Amazon search results, creating a potential surface for indirect prompt injection. * Ingestion points: External data enters the agent context via the call_api function in scripts/amazon_search.py, which retrieves real-time product listings from a remote API. * Boundary markers: No specific delimiters or safety instructions are provided to the agent to distinguish between valid tool output and potential embedded instructions within product titles or descriptions. * Capability inventory: The script has permissions to write result files to the local filesystem (under the nexscope directory) and provide summaries of that data to the agent's context. * Sanitization: The skill does not perform sanitization or filtering of external product strings before they are displayed or saved.

ecommerce-amazon-search