The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected in the skill instructions or metadata.
[EXTERNAL_DOWNLOADS]: The skill is configured to fetch documentation from learn.microsoft.com. These are official sources from a well-known service and are consistent with the skill's purpose as an Azure documentation assistant.
[PROMPT_INJECTION]: Instructions provided to the agent are standard operational guidelines for handling large files (line range reading) and checking for version updates. No bypass or override commands were found.
[CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or sensitive credentials were found in the file content or metadata.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: Fetches external documentation via mcp_microsoftdocs:microsoft_docs_fetch and fetch_webpage tools.
Boundary markers: None explicitly defined in the provided file for the tool outputs.
Capability inventory: The skill contains no executable scripts, subprocess calls, or file-write operations. It is a pure documentation reference skill.
Sanitization: Not applicable as no local data processing or interpolation is performed.

azure-ai-services