mcp-configure

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses the Azure CLI (az) to verify authentication and obtain access tokens for environment discovery.
  • [COMMAND_EXECUTION]: Creates local directories for project-scoped configuration files using system commands.
  • [DATA_EXFILTRATION]: Performs network requests to well-known Microsoft API endpoints (api.powerapps.com) to discover Dataverse environments.
  • [DATA_EXFILTRATION]: Accesses the agent's local configuration files (~/.copilot/mcp-config.json and .mcp/copilot/mcp.json) to read existing settings and register new MCP servers.
  • [PROMPT_INJECTION]: Processes Dataverse environment names from a remote API, creating a surface for indirect prompt injection.
  • Ingestion points: Environment metadata from api.powerapps.com.
  • Boundary markers: None used during display.
  • Capability inventory: File system read/write access and Azure CLI command execution.
  • Sanitization: Basic URL normalization; environment display names are not sanitized.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 03:14 PM