The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (YugabyteDB Anywhere API and Prometheus instances) which provides a surface for indirect prompt injection if those sources return malicious content.
Ingestion points: API responses are ingested via yba_request in references/python-client.md and Invoke-YbaRequest in references/powershell-client.md. Metric data is ingested via prom_range in references/prometheus.md.
Boundary markers: None. The skill does not provide instructions to the agent to treat API response content as untrusted data or use delimiters.
Capability inventory: The skill has network access capabilities via the requests library in Python and Invoke-RestMethod in PowerShell to perform GET, POST, PUT, and DELETE operations.
Sanitization: There is no evidence of sanitization or filtering of the content returned by the API or Prometheus before it is processed by the agent.
[COMMAND_EXECUTION]: The skill provides instructions and examples for executing shell commands to manage network access and query services.
Evidence: references/prometheus.md contains instructions for using kubectl port-forward to access internal Prometheus instances and curl for querying metrics.

yba-api