device-guidance-v1-0-1
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill processes and displays data retrieved from the
Get Employee Assetstool. This creates a potential surface where the data provided by the tool could influence the agent's behavior if that source contains unexpected instructions. - Ingestion points: The skill ingests untrusted data from the output of the
Get Employee Assetstool as defined in the 'Look up available options' section ofSKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions for the agent to treat the tool output strictly as data rather than instructions.
- Capability inventory: The agent is capable of formatting tool data into tables, rendering external image URLs, and using SharePoint Item IDs to perform follow-up request actions.
- Sanitization: There are no explicit sanitization or validation steps mentioned for the fields (such as 'Key details' or 'Image URL') returned by the asset tool before they are presented to the user.
Audit Metadata