The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, such as business requirements and existing CDS schema files, which serves as a potential attack surface for indirect instructions.
Ingestion points: The agent processes user-provided business descriptions and reviews local schema.cds files as part of its core workflow defined in agents/cds-modeler.md.
Boundary markers: The instructions do not define specific delimiters or strict guidance to help the agent distinguish between data and potentially malicious commands embedded within the processed content.
Capability inventory: The skill's primary function is generating and reviewing CDS code; it does not request or utilize tools for network access, file system modifications, or command execution.
Sanitization: There are no explicit instructions for the agent to sanitize, filter, or validate the content of the ingested files before processing them.

cds-modeling