The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill correctly identifies and addresses performance issues within its stated technical scope.
[COMMAND_EXECUTION]: The skill utilizes specialized Power BI tools such as dax_query_operations, model_operations, and trace_operations to analyze technical metadata and measure query performance. These operations are restricted to the Power BI semantic model environment and are necessary for the skill's diagnostic functions.
[EXTERNAL_DOWNLOADS]: The skill uses the microsoft_docs_search tool to fetch guidance from Microsoft Learn. This is a reference to a well-known, trusted technical service for official documentation and does not involve executing untrusted remote code.
[PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection as it ingests Power BI model metadata and DAX queries (ingestion points in SKILL.md and referenced files). No explicit boundary markers or sanitization logic are defined in the instructions, and the capability inventory includes various model and query operations across all scripts. However, there is no evidence of malicious intent or bypass patterns, and the risk is assessed as minimal given the technical troubleshooting context.

power-bi-performance-troubleshooting