The Agent Skills Directory

[SAFE]: The skill serves as a structured reference for Azure Automation development. All external resources and documentation links point to learn.microsoft.com, which is an official and trusted domain.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch external content from specified Microsoft documentation URLs. While fetching external data introduces a theoretical ingestion surface for indirect prompt injection, the risk is mitigated by the use of a highly trusted source and the absence of dangerous system capabilities (like arbitrary file writes or command execution) within the skill's scope.
Ingestion points: External documentation fetched via mcp_microsoftdocs:microsoft_docs_fetch or fetch_webpage (SKILL.md).
Boundary markers: None explicitly defined in the skill instructions.
Capability inventory: No file system write, subprocess execution, or arbitrary network operations detected.
Sanitization: None specified in the skill body.
[EXTERNAL_DOWNLOADS]: The skill references an installation guide for the mcp_microsoftdocs tool on GitHub. This repository (github.com/MicrosoftDocs/mcp) is owned by a trusted organization and is used for legitimate tool configuration.

azure-automation