The Agent Skills Directory

[SAFE]: The skill provides integration for Databricks Genie, allowing for the creation and querying of natural language SQL spaces within the Databricks ecosystem. Analysis of the instructions and documentation across SKILL.md, conversation.md, and spaces.md reveals no malicious patterns.
[COMMAND_EXECUTION]: The skill uses tools like execute_sql and ask_genie. These are the primary, intended functions of the skill for data exploration and management within the Databricks SQL warehouse environment.
[EXTERNAL_DOWNLOADS]: No external downloads or remote script executions from non-trusted or unknown sources were detected. References to other Databricks-related skills are internal to the vendor's ecosystem.
[CREDENTIALS_UNSAFE]: No hardcoded credentials, API keys, or sensitive file paths were found. The skill relies on environment-provided authentication for the Databricks platform.
[INDIRECT_PROMPT_INJECTION]: The ask_genie tool processes natural language input to generate SQL. While this is a surface for prompt injection, it is the fundamental purpose of the 'Genie' service, which includes its own internal safeguards for SQL generation.

databricks-genie