The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and display content from Amazon product pages, including product descriptions, bullet points, and customer reviews. This external data is ingested into the agent's context and could potentially contain malicious instructions or "jailbreak" attempts hidden in listing text.
Ingestion points: scripts/amazon_product_detail.py fetches product data from the NexScope API, which the agent then processes.
Boundary markers: The skill does not define specific delimiters or warnings to the agent to ignore instructions embedded within the product data.
Capability inventory: The skill uses a Python script for network communication and writes response data to the local file system.
Sanitization: There is no evidence of filtering or sanitizing the retrieved Amazon listing content before it is presented to the agent.
[COMMAND_EXECUTION]: The skill utilizes a Python script (scripts/amazon_product_detail.py) to interface with an external API. The agent is instructed to execute this script with JSON-formatted arguments to retrieve data.
[CREDENTIALS_UNSAFE]: The script and documentation direct the agent to access the NEXSCOPE_API_KEY environment variable for authentication. While using environment variables is a recommended practice for secret management, the skill's functionality depends on accessing this sensitive information.

ecommerce-amazon-product-detail