The Agent Skills Directory

[SAFE]: Documentation Retrieval: The skill is configured to fetch technical content from learn.microsoft.com. This is an official, well-known, and trusted domain for Microsoft documentation. The use of a query parameter from=learn-agent-skill is a benign telemetry practice.
[SAFE]: Trusted Tooling: The skill recommends the use of mcp_microsoftdocs, a tool hosted in an official Microsoft GitHub organization (MicrosoftDocs). This represents a secure chain of trust between the skill author and the recommended extensions.
[SAFE]: Indirect Prompt Injection Surface: While the skill ingests external data from the web (Category 8 surface), the target source is a restricted, authoritative vendor documentation site, which significantly mitigates the risk of adversarial content injection.

azure-reliability