Overview
The Conveyor MCP Server lets AI assistants — like Claude, ChatGPT, Cursor, and others — interact directly with your Conveyor account through the Model Context Protocol (MCP). Instead of switching between your AI tool and the Conveyor dashboard, you can ask your assistant to manage questionnaires, update your Trust Center, approve access requests, and pull analytics - all through natural conversation.
MCP is an open standard that gives AI models a structured way to call external tools. Conveyor's MCP server exposes your Conveyor API as a set of tools that any MCP-compatible client can discover and use automatically.
What can I do with it?Here are a few examples of things you can ask your AI assistant once connected:
- "What questionnaires are due this week?"
- "Approve all pending access requests from @acme.com"
- "Do you encrypt data at rest?"
- "Which customers downloaded our SOC 2 report in the last 30 days?"
🛠️ Available Tools
Once connected, your AI assistant will automatically discover all available tools. These tools map directly to Conveyor's public API — the MCP server provides the same capabilities, just accessible through natural language instead of raw API calls.
🔑 Authentication
| Mode | How it works | Best for |
|---|---|---|
| OAuth 2.1 (preferred) | You click "Connect" in your AI client, log into Conveyor in the browser, pick which organization to authorize. Permissions match your Conveyor user. | Claude, ChatGPT, any modern MCP client |
| API Token (Bearer) | Generate an API token in Conveyor - see Generating an API Token for instructions. Paste the token into your client's MCP config (e.g. mcp.json). Permissions are tied to the token's grants. | Cursor, OpenAI Playground, MCP Inspector |
What if my client supports both?Prefer OAuth. It's tied to your Conveyor user (so revoking happens automatically when your account is deactivated or your password rotates) and it inherits the same per-organization permissions you have in the dashboard.
⚡ Initial MCP Server Setup
Option 1 - OAuth (Claude, ChatGPT, any modern MCP client)
For OAuth-capable clients, add the below server URL as a custom connector / app. The client handles the rest of the flow.
https://mcp.conveyor.com/mcp
Enterprise accountsSome AI providers (ChatGPT Business, Claude Enterprise, etc.) require an organization admin to register custom connectors before individual users can use them. If Conveyor doesn't appear in your connector list after following these steps, ask your IT admin to add it once for the whole organization.

Adding the Conveyor MCP server to Claude Desktop as a custom connector

Adding the Conveyor MCP server to ChatGPT Business as a custom app
Option 2 - OpenAI Playground
Any MCP-compatible client can connect by pointing to the server URL and including your API token as a Bearer token in the Authorization header:
| Setting | Value |
|---|---|
| Server URL | https://mcp.conveyor.com/mcp |
| Auth type | Bearer token |
| Token | Your Conveyor API token |
Refer to your client's documentation for where to configure remote MCP servers.

Adding the Conveyor MCP Server to the OpenAI Playground using API key authentication
Option 3 - JSON Configured Clients (Cursor, Claude Code, etc)
Add the following configuration to the client's JSON settings:
{
"mcpServers": {
"conveyor": {
"url": "https://mcp.conveyor.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_TOKEN"
}
}
}
}For Claude Code, you may also run the following command to add Conveyor to your MCP server list
claude mcp add --transport http Conveyor https://mcp.conveyor.com/mcp --header "Authorization: Bearer YOUR_API_TOKEN"
Connecting to the MCP Server
Enterprise accountsSome AI providers (ChatGPT Business, Claude Enterprise, etc.) require an organization admin to register custom connectors before individual users can use them. If Conveyor doesn't appear in your connector list after following these steps, ask your IT admin to add it once for the whole organization.
Claude
Step 1: Select the "+" icon -> "Manage connectors"

Step 2: Select the Conveyor MCP server and hit "Connect"

ChatGPT Business
Step 1: Search for Conveyor in the app list

Step 2: Select "Connect"

💡 Example Workflows
Here are a few ways to get the most out of the MCP server by combining multiple tools in a single conversation:
Ask one off security questions
"Do you encrypt data at rest?"
Weekly questionnaire check-in
"Show me all questionnaires that are due in the next 7 days, grouped by status. Flag any that are still in 'started' and haven't moved to review yet."
Customer engagement report
"Pull this month's Trust Center interactions and cross-reference with our connections list. Who are our top 5 most engaged customers, and what are they downloading?"
Bulk access management
"Find all pending access requests from @stripe.com email addresses, approve them, and add them to the Enterprise access group."
Knowledge Base audit
"Which Knowledge Base questions are getting the most views this quarter? Are any of them in a status that needs review?"
❓ FAQ
Do I need to self-host anything? No. Conveyor hosts the MCP server at mcp.conveyor.com/mcp. Just point your client to it and authenticate via OAuth or API token.
Should I use OAuth or an API token? OAuth if your client supports it — it's tied to your Conveyor user (auto-revoked when your password rotates or your account is deactivated) and there's no long-lived bearer token sitting in a config file. API tokens for everything else.
Can the AI assistant access data outside my account? No. In the case of OAuth authentication, the token inherits the same permissions as the logged in user. In the case of API token authentication, the assistant can only access data your token is authorized to see.
Can I revoke OAuth access? Yes. You can revoke a specific OAuth session from your Conveyor dashboard, or revoke all credentials by changing your password (which automatically invalidates any OAuth tokens issued under the old credential).
Does Conveyor see my AI conversation? No. The AI client running on your machine (or in your browser) talks to Conveyor only when it explicitly calls an MCP tool. Your conversation with the AI assistant stays between you and the AI provider.

