power-bi-dax-development
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading existing measures and model metadata using the
measure_operationstool to inform the development of new DAX measures. This ingestion of potentially untrusted metadata (such as measure names or descriptions provided by other users or processes) creates a surface for indirect prompt injection if those fields contain malicious instructions designed to influence the agent's behavior during the 'Research' or 'Write DAX' phases. - Ingestion points: Output from the
measure_operationstool (specifically when listing current measures in Step 2 of the workflow). - Boundary markers: The skill does not define specific delimiters or instructions to treat existing model metadata as untrusted data.
- Capability inventory: The agent has the capability to create new measures via
measure_operationsand execute arbitrary DAX queries usingdax_query_operations, which could be exploited if an injection is successful. - Sanitization: No sanitization or validation of existing metadata is described before it is incorporated into the agent's context.
Audit Metadata