ado-init

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill configures the agent to use the @azure-devops/mcp package via npx. This package is an official tool provided by Microsoft, which is a trusted organization. This reference is documented neutrally as it targets a well-known, trusted source.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the configuration of an MCP (Model Context Protocol) server by writing execution commands (npx -y @azure-devops/mcp) to a .mcp.json file. This is the intended primary function of the skill to enable Azure DevOps integration.
  • [COMMAND_EXECUTION]: The skill generates platform-specific command strings (using cmd /c npx for Windows and npx for Linux/Mac) to be used in the local environment configuration.
  • [PROMPT_INJECTION]: The instructions include defensive constraints requiring the agent to ignore any initial user arguments and strictly follow an interactive multi-phase prompt sequence. This is a logic-flow enforcement mechanism rather than a malicious bypass of safety guidelines.
  • [DATA_EXPOSURE]: The skill implements a security-conscious approach to configuration management. It explicitly forbids the agent from reading the contents of an existing .mcp.json file to prevent accidental exposure of other configured service credentials, instead providing a manual update snippet to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 12:48 PM