The Agent Skills Directory

[COMMAND_EXECUTION]: Uses python3 and npx to execute commands for network topology discovery and Microsoft Graph API operations.
[EXTERNAL_DOWNLOADS]: Downloads the official @anthropic-ai/microsoft-graph-mcp package from the npm registry at runtime using npx.
[CREDENTIALS_UNSAFE]: Requires Azure Active Directory credentials (AZURE_CLIENT_ID, AZURE_CLIENT_SECRET) passed via environment variables for SharePoint authentication. No secrets are hardcoded in the skill code.
[PROMPT_INJECTION]: Features an indirect prompt injection surface where the agent processes data received from network devices.
Ingestion points: Reads device names and neighbor information from show cdp neighbors detail and show lldp neighbors detail outputs in SKILL.md.
Boundary markers: The workflow does not specify the use of delimiters or 'ignore' instructions when processing discovery output.
Capability inventory: The skill can write files to SharePoint and execute administrative commands via the Microsoft Graph MCP server.
Sanitization: No validation or sanitization logic is described for the incoming network discovery data before it is formatted into Visio OOXML or Mermaid content.

msgraph-visio