The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes DNS record sets and zone metadata that are externally controlled, creating a potential surface for indirect prompt injection attacks.
Ingestion points: Data is retrieved from the GCP environment via gcloud dns managed-zones list and gcloud dns record-sets list within SKILL.md.
Boundary markers: The skill does not implement delimiters or explicit warnings to the agent to ignore instructions potentially embedded within the DNS record content.
Capability inventory: The skill utilizes gcloud CLI subprocess calls for listing and describing DNS resources as defined in SKILL.md.
Sanitization: No filtering or sanitization of record values is performed before the data is returned to the agent's context.

managing-google-cloud-dns