RAIC MCP Server
The RAIC MCP Server lets AI assistants query your governance data over the Model Context Protocol. An assistant can answer questions about your AI systems, controls, risks, and policies directly — under OAuth authorization, scoped permissions, and full audit logging.
What MCP is
Model Context Protocol is an open standard for connecting AI assistants to external systems. Instead of pasting exports into a chat window, the assistant calls a defined set of tools against a server that enforces authentication and authorization. RAIC exposes such a server so assistants can read governance data without any human copying it out of the platform.
Connecting a client
- Server address to paste into your client: https://app.rhindoncyber.com/api/public/raic-mcp
- Before anyone connects, an Org Admin switches MCP on under Settings, Integrations, MCP Server. Until then every tool call is refused.
- Supported clients: ChatGPT (Connectors), Claude Desktop (through the mcp-remote adapter), Cursor, Codex CLI, Perplexity, and Microsoft 365 Copilot through the Ask RAIC agent.
- Each client registers itself and sends you to RAIC to sign in; there is no client ID or secret to copy.
- Ask RAIC is a Microsoft 365 Copilot agent your Microsoft 365 administrator installs from the RAIC app package. It uses this same server and the same consent screen.
Authorization flow
- The MCP client registers and initiates an OAuth 2.1 authorization request against RAIC.
- You sign in and are shown a consent screen naming the client and listing the RAIC permissions it can have.
- On approval, the client receives a short-lived access token bound to your user and organization.
- Every tool call is executed as you, in your organization, subject to the same row-level security as the web interface.
- Tokens refresh on the standard OAuth cycle and can be revoked at any time; revocation takes effect immediately.
Tool surface
The exposed tools are a fixed allowlist, each one a call to the RAIC REST API, generated from its OpenAPI spec and named after the method and path (for example get_v1_risks). There is no arbitrary query capability: an assistant cannot construct SQL or reach a table that is not behind a tool. Tools that change data run only when your organization's MCP policy is read-write and you granted the matching write permission. Tools cover AI systems and use cases, controls, the risk register, suppliers, governance policies, maturity scores, phishing simulations, HIPAA registers when the HIPAA module is on, the audit log, the Getting Started checklist (for Org Admins), and adding Partner clients (for Partner admins). Endpoint details are in the RAIC REST API Reference at /support/raic-api.
Scopes and least privilege
- You choose the RAIC permissions on the consent screen. Read permissions are ticked for you; write permissions are off until you tick them.
- If you approve without choosing any RAIC permission, the connection is read-only.
- A scope can only narrow what your own role already permits — MCP never elevates access.
- Org boundaries are enforced server-side; a token issued for one organization cannot read another.
- Writes also need your organization's MCP policy set to read-write by an Org Admin; under the default read-only policy every write tool is refused.
- The assistant still sees every tool, but a call that needs a permission you declined is refused with insufficient_scope, and the refusal is audit-logged.
Audit coverage
Every MCP authorization, token issuance, revocation, and tool invocation writes an audit entry scoped to your organization and mirrored to the platform audit log. Entries record the client, the user, the tool called, and the outcome. Failed authorization attempts are logged too, with the token identifier hashed rather than stored, so you can spot a misconfigured or stale client without exposing credentials.
Administration
- The MCP Admin Console lists registered clients, active grants, and recent tool activity.
- Revoke a client's grant to cut off access without disturbing other integrations.
- Per-client activity view helps you attribute unexpected query volume.
- Access to the console is restricted to administrators.
Good practice
Treat an MCP grant like any other credential: authorize only clients you control, review grants periodically, and revoke ones you no longer use. Because the assistant sees real governance data, apply the same care to where its conversation history is stored as you would to an exported report.
