The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from the Sage Intacct REST API and forwards it to the agent, creating a standard surface for indirect prompt injection.
Ingestion points: The skill processes requests and responses from the Sage Intacct REST API via the Proxy API (referenced in SKILL.md).
Capability inventory: The skill utilizes network operations via https://unify.apideck.com/proxy to communicate with external endpoints.
Boundary markers: The instructions do not specify boundary markers or specific delimiters for data ingested from the external API.
Sanitization: There is no mention of sanitization or validation routines for data received from the downstream Sage Intacct service.
[SAFE]: The skill utilizes legitimate vendor-owned infrastructure for its core operations.
All network operations are directed to unify.apideck.com, which is the official production environment for Apideck services.
The instructions explicitly follow best practices by directing the agent to use managed Vault authentication instead of asking the user for sensitive credentials.

sage-intacct-rest