microsoft-docs
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including Microsoft documentation and the SandBase API, which creates a potential surface for indirect prompt injection. If the retrieved content contains malicious instructions, the agent might attempt to execute them.
- Ingestion points: External data is ingested through the sandbase_run and sandbase_run_get tools, as well as general research of Microsoft Docs.
- Boundary markers: The instructions do not define explicit delimiters or 'ignore instructions' markers to separate untrusted external content from the core logic.
- Capability inventory: The skill utilizes tool calls for discovery, inspection, and execution of external capabilities via the vendor's API.
- Sanitization: There is no mention of explicit sanitization, filtering, or validation of the content retrieved from external sources before it is processed by the agent.
Audit Metadata