The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to ingest and analyze application logs, which represents a potential attack surface for indirect prompt injection if the logs contain adversarial content.
Ingestion points: SKILL.md describes retrieving logs via the databricks apps logs CLI command and the /logz browser endpoint.
Boundary markers: The instructions do not specify the use of delimiters or provide warnings to the agent to disregard instructions embedded within the log data.
Capability inventory: The skill instructions include the use of the databricks CLI and jq for processing application status and log streams.
Sanitization: There are no instructions for sanitizing or escaping the log content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill provides instructions for using standard command-line tools and system signals to debug applications within the Databricks environment.
Evidence includes usage of databricks apps logs, databricks apps get, and the kill -USR1 command to trigger thread dumps. These are legitimate diagnostic actions provided by the vendor for application management.

diagnostics