The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from the Amazon Seller API, including customer comments via the sales-intel returns command.
Ingestion points: Data enters the agent context through API responses for listing details, order records, and specifically customer comments in returns analysis.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the fetched data.
Capability inventory: The skill allows the agent to execute shell commands, write output to local files, and POST data to external webhooks.
Sanitization: There is no evidence of sanitization or filtering applied to external content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the amazon-seller-pp-cli binary from the vendor's infrastructure.
Evidence: It uses npx -y @mvanhorn/printing-press-library install amazon-seller and go install github.com/mvanhorn/printing-press-library/... to set up the necessary tools.
[COMMAND_EXECUTION]: The skill is designed to execute local shell commands via the amazon-seller-pp-cli tool to manage inventory, reports, and seller account data.
[DATA_EXFILTRATION]: The Output Delivery feature provides a mechanism (--deliver webhook:<url>) to transmit command results directly to an external URL.
Evidence: The skill documentation describes a webhook:<url> sink that POSTs the output body to a specified endpoint.
[CREDENTIALS_UNSAFE]: The skill instructs the user to export sensitive API credentials as environment variables for the tool to function.
Evidence: Users are prompted to export SP_API_LWA_CLIENT_ID, SP_API_LWA_CLIENT_SECRET, and SP_API_REFRESH_TOKEN.

pp-amazon-seller