The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that processes and transmits user-supplied personal information and configuration data to external service endpoints, creating an interface for indirect data processing.
Ingestion points: Data parameters such as phoneNumber, name, senderId, firstName, lastName, streetAddress, locality, postalCode, countryCode, documentType, and storageId are ingested through SDK calls and REST API commands in SKILL.md.
Boundary markers: The instructions do not specify any delimiters or safety prompts to isolate user-provided content or warn the agent about potentially malicious instructions embedded in that data.
Capability inventory: The skill utilizes an SDK (zavu), the fetch API for document uploads, and curl for interacting with the api.zavu.dev service.
Sanitization: There are no explicit requirements or logic defined for the agent to sanitize or validate the user-provided inputs before they are transmitted to the API.

phone-numbers