The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill is designed to fetch remote documentation from learn.microsoft.com and references an installation guide for MCP tools on github.com/MicrosoftDocs. Both sources are well-known services and belong to a trusted organization, representing legitimate vendor functionality.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it incorporates external content directly into the agent's operating context.
Ingestion points: Remote documentation content is fetched from learn.microsoft.com using the microsoft_docs_fetch or fetch_webpage tools as described in the compatibility and category index sections of SKILL.md.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings to isolate the fetched documentation from the agent's core instructions.
Capability inventory: The skill utilizes read_file for local section access and relies on network-enabled tools (microsoft_docs_fetch, fetch_webpage) to retrieve external data.
Sanitization: The skill lacks specified logic for sanitizing, escaping, or validating the Markdown content retrieved from external URLs before processing it.

azure-networking