The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses various command-line tools including agents-cli, gcloud, terraform, and kubectl to perform deployment and infrastructure management tasks. These operations are core to the skill's purpose and include safeguards such as requiring explicit human approval for deployments.\n- [EXTERNAL_DOWNLOADS]: Fetches deployment guidelines and documentation from adk.dev. It also references official samples from the google/adk-samples GitHub repository. These sources are well-known and directly related to the supported development kit.\n- [INDIRECT_PROMPT_INJECTION]: The implementation guide for BigQuery batch inference in references/batch-inference.md introduces a potential injection surface.\n
Ingestion points: The trigger_bq endpoint processes data directly from BigQuery table rows passed via the calls parameter.\n
Boundary markers: The reference code does not demonstrate the use of delimiters or instructions to compartmentalize ingested data from the agent's core instructions.\n
Capability inventory: The skill has the capability to run agents and trigger deployments, though deployment requires separate authorization.\n
Sanitization: Input strings are serialized to JSON but lack specific sanitization or safety filtering to prevent malicious instructions embedded in BigQuery rows from affecting agent behavior.

google-agents-cli-deploy