anfra-lookup-values
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the
anfraCLI tool to query distinct values from a dataset. This is the primary intended function of the skill for data discovery within the Holistics ecosystem. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect injection as it processes data returned from the
anfra querycommand (the database's distinct values). - Ingestion points: Data enters the context from the output of the
anfra querycommand in the shell. - Boundary markers: No explicit boundary markers or 'ignore' instructions are provided for the agent when processing the query results.
- Capability inventory: The skill uses
anfra query(command execution) to interact with the database. - Sanitization: There is no evidence of sanitization for the data returned from the database or for the user-supplied patterns interpolated into the AQL (Anfra Query Language) strings.
Audit Metadata