The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from Shopify (such as order details, product descriptions, and customer names), which acts as an untrusted input source that could potentially contain malicious instructions targeting the agent.
Ingestion points: Data is retrieved through Apideck Ecommerce unified API resources including orders, products, customers, and stores as detailed in SKILL.md.
Boundary markers: The instructional examples do not include explicit boundary markers or instructions for the agent to ignore potential commands embedded within the retrieved ecommerce data.
Capability inventory: The skill enables significant capabilities including listing orders (orders.list) and creating new products or customers (products.create, customers.create), providing a write-back surface for processed data.
Sanitization: There is no evidence of data sanitization, filtering, or validation logic in the provided TypeScript and Bash examples to mitigate risks from processing external content.

shopify