The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Printify API including shop details, product listings, and order information.
Ingestion points: Data is fetched via printify shops, products, orders, and catalog lookup commands.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat data from these API calls as untrusted content.
Capability inventory: The agent has the ability to write or delete data, including create-product, update-product, delete-product, and create-order.
Sanitization: No sanitization or validation of the ingested API content is specified before it is processed by the agent.
[COMMAND_EXECUTION]: The skill makes extensive use of a local printify CLI tool to interact with external services. While the commands are structured and restricted to the Printify domain, the execution of these commands depends on the agent correctly interpreting user intent and API responses.
[DATA_EXFILTRATION]: The upload --file <path> command allows the agent to read local files and transmit them to Printify's servers. While this is the primary purpose for design uploads, it presents a potential vector for sensitive file exfiltration if the agent is manipulated into targeting system files (e.g., credentials or configuration files) instead of legitimate image assets.

printify