The Agent Skills Directory

[SAFE]: The skill serves as an expert knowledge base and reference guide for Azure Virtual WAN development, troubleshooting, and architecture. It contains structured information to guide an AI agent.
[EXTERNAL_DOWNLOADS]: The skill references a large number of official technical articles from learn.microsoft.com. Fetches documentation from Microsoft's official domains, which is a well-known and trusted service. These references are documented neutrally as they represent the core purpose of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it instructs the agent to ingest content from external documentation URLs.
Ingestion points: Documentation fetched via mcp_microsoftdocs:microsoft_docs_fetch or WebFetch as described in SKILL.md.
Boundary markers: None provided in the skill instructions.
Capability inventory: The skill contains no executable code, scripts, or privileged commands. It is purely informational.
Sanitization: No sanitization logic is present for the fetched content.
[NO_CODE]: The skill does not include any scripts, automation, or binaries. It consists entirely of Markdown instructions and metadata, which significantly reduces the potential for malicious behavior.

azure-virtual-wan