About MCPForge Scores MCPForge Scores are proprietary assessment metrics designed to help identify potential security, compliance, compatibility, quality and operational risks. Scores should be treated as guidance and not as a formal certification, audit or compliance determination.
90
Security
How is this calculated?
✓Authentication requirements
✓Destructive operations
✓Sensitive endpoints
✓Approval workflows
✓Tool permissions
✓Audit coverage
This score is an MCPForge risk assessment metric and should be treated as guidance rather than a formal security audit.
For this server
+Server reachable and responding
−5 destructive operations detected
−Credentials Vault not configured — API keys may be exposed through direct agent access if tools require authentication
82
Compliance
How is this calculated?
✓PII handling
✓Customer data exposure
✓Billing operations
✓User deletion operations
✓Regulatory-sensitive endpoints
This score indicates potential compliance considerations and is not legal advice.
For this server
−User/contact data handling detected (GDPR/CCPA obligations)
−Credentials Vault not configured — configure to ensure proper server-side key management
90
Compat
How is this calculated?
✓MCP protocol support
✓Tool naming
✓Schema quality
✓Claude Desktop compatibility
✓Resources support
✓Prompts support
For this server
+MCP protocol handshake successful
+116 tools discovered
+All tool names compatible with Claude Desktop
+resources/list endpoint supported
+prompts/list endpoint supported
−116 tools — large count may impact agent context
90
Quality
How is this calculated?
✓Pagination support
✓Filtering support
✓Bounded responses
✓Tool descriptions
✓Parameter quality
✓API design characteristics
For this server
+All tools have descriptions
95
Health
How is this calculated?
✓Success rate
✓Error rate
✓Response latency
✓Tool availability
✓Execution reliability
For this server
+All protocol tests passed
+Tool execution verified successfully
+Acceptable latency (952ms)
+116 tools operational
Overview
Microsoft Graph MCP gives AI agents access to messaging — channels, messages, and users. Built with MCPForge, it includes audit logging, tool-level permissions, and governance controls.