The Agent Skills Directory

[PROMPT_INJECTION]: The skill allows the agent to read and process attachment data from ServiceNow, which represents an indirect prompt injection surface.
Ingestion points: The skill reads attachment content through functions like getAttachmentContent and retrieves metadata via snow_query_table on the sys_attachment_doc table.
Boundary markers: No instructions are provided to isolate or treat attachment content as untrusted data within the prompt context.
Capability inventory: The skill provides access to the snow_execute_script_with_output tool, which enables the agent to execute arbitrary server-side code on the ServiceNow instance.
Sanitization: The skill lacks sanitization or validation logic for the content retrieved from ServiceNow attachment records before it is processed.

document-management