m365
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple shell commands using the
m365CLI to interact with Microsoft Graph API and standard shell utilities likecatfor file creation. - [EXTERNAL_DOWNLOADS]: The
setup.mdinstructions include a global installation of the CLI for Microsoft 365 (npm install -g @pnp/cli-microsoft365) from the npm registry. - [INDIRECT_PROMPT_INJECTION]: The skill reads external data from email messages and calendar events (SKILL.md). This ingested data is not sanitized and does not use boundary markers. The skill has significant capabilities, including sending emails and modifying calendar events, creating a surface for malicious data to influence agent actions.
- [CREDENTIALS_UNSAFE]: The setup process requires the user to provide and enter a Client Secret via a CLI flag (
--secret). This practice exposes sensitive credentials in the process list and potentially in shell history files.
Audit Metadata