The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches documentation content from official Microsoft domains (learn.microsoft.com). This is an expected and transparent behavior given the skill's purpose, and the target is a well-known, trusted service.
[INDIRECT_PROMPT_INJECTION]: By fetching external documentation at runtime, the skill exposes a surface for indirect prompt injection. While the sources are trusted, the skill does not explicitly define boundary markers or instruct the agent to disregard instructions potentially embedded in the fetched documentation content.
[SAFE]: No evidence of prompt injection, credential exfiltration, obfuscation, or persistent access was found. The skill's implementation is consistent with its stated goal of providing security configuration expertise.

azure-security