Skip to content
Dipole Docs

MCP Identity Gateway

The MCP Identity Gateway sits between AI agents and MCP servers, enforcing policies and injecting per-user credentials.

Two-Policy Model

Section titled “Two-Policy Model”

The Gateway requires two Access Policies:

Policy 1: Client-to-Gateway

Section titled “Policy 1: Client-to-Gateway”
  • Client Workload — AI agent identified by redirect URI
  • Server Workload — The Gateway endpoint
  • Trust Provider — OIDC token validating the human user
  • Credential Provider — OIDC token embedding user identity

Policy 2: Gateway-to-Server

Section titled “Policy 2: Gateway-to-Server”
  • Client Workload — The Gateway itself
  • Server Workload — The actual MCP server
  • Trust Provider — Validates the Gateway’s token
  • Credential Provider — Per-user OAuth token for the MCP server

Per-User Credential Isolation

Section titled “Per-User Credential Isolation”

Each user gets their own MCP server credentials:

  1. User authenticates via IdP when connecting through the Gateway
  2. First-time access triggers OAuth consent with the MCP server
  3. Dipole stores each user’s token separately
  4. Subsequent requests inject that user’s specific token
  5. Audit trail shows exactly which user accessed which tool