The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script fetch_product.sh utilizes Python heredocs to dynamically construct API request payloads and validate response data at runtime.
[COMMAND_EXECUTION]: The tool provides a mechanism to write API output to an arbitrary file path via the --output flag, which involves executing mkdir and cp commands on the destination path.
[INDIRECT_PROMPT_INJECTION]: The skill ingests product data from Amazon, which could contain adversarial content designed to influence the agent's behavior during data summarization.
Ingestion points: Amazon product details retrieved from the Frevana API in scripts/fetch_product.sh.
Capability inventory: The skill has the ability to perform network requests and write results to the local file system.
Boundary markers: No delimiters or explicit instructions are used to isolate the untrusted product data from the agent's instructions.
Sanitization: The script validates that the response is JSON but does not perform content-level sanitization or filtering of the retrieved text.

amazon-product