email-manager

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Microsoft 365 email messages, which creates a potential attack surface.
  • Ingestion points: The skill uses mcp__ms365__list-mail-messages and mcp__ms365__get-mail-message to retrieve email headers and body content into the agent's context.
  • Boundary markers: The instructions lack explicit delimiters or instructions to ignore embedded commands when the agent analyzes or summarizes email content.
  • Capability inventory: The skill provides tools to modify the mailbox, including mcp__ms365__move-mail-message, mcp__ms365__delete-mail-message, and mcp__ms365__send-mail for forwarding messages.
  • Sanitization: There is no evidence of content sanitization or filtering before the agent processes the email data.
  • [EXTERNAL_DOWNLOADS]: The skill documentation describes how to fetch the required integration package.
  • The configuration instructions specify using npx -y @anthropic-ai/mcp-server-ms365@latest to install the Microsoft 365 MCP server from a well-known public registry.
  • [COMMAND_EXECUTION]: The skill relies on a command-line tool to perform its operations.
  • The integration utilizes the npx command to execute the official Microsoft 365 MCP server as part of its standard workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 07:38 PM