Cursor Setup

Connect SanctumAI to Cursor so your AI coding assistant can securely access credentials.

Prerequisites

• SanctumAI CLI installed (Installation guide)
• Vault initialized with credentials
• Cursor installed

Step 1: Register an Agent

sanctum agent register cursor-agent --description "Cursor AI editor"

Step 2: Create a Policy

sanctum policy add cursor-access \
  --principal "agent:cursor-agent" \
  --resources "openai/*,anthropic/*" \
  --actions retrieve \
  --max-ttl 300

Step 3: Configure Cursor

Create .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "sanctum": {
      "command": "sanctum",
      "args": ["mcp", "serve"]
    }
  }
}

Note: This is a per-project configuration. To use SanctumAI across all projects, create the file in your home directory or use Cursor’s global MCP settings.

Step 4: Start the Daemon

sanctum daemon start

Step 5: Restart Cursor

Restart Cursor (Cmd+Shift+P → “Reload Window” or quit and reopen).

Go to Settings → MCP and verify “sanctum” appears in the server list with a green status indicator.

Step 6: Test

In Cursor’s AI chat, ask:

“Use Sanctum to retrieve my OpenAI API key”

Cursor’s agent will:

1. Call sanctum_get_credential via MCP
2. Authenticate as cursor-agent
3. Receive a time-limited lease
4. Use the credential

Using Credentials in Code Generation

You can also ask Cursor to use Sanctum when generating code:

“Create a Python script that calls the OpenAI API. Get the API key from Sanctum instead of using an environment variable.”

Troubleshooting

Sanctum not appearing in MCP servers

1. Make sure the daemon is running: sanctum daemon status
2. Check that .cursor/mcp.json is valid JSON
3. Restart Cursor completely (not just reload window)

“Access denied” errors

Run a policy simulation:

sanctum policy simulate --agent cursor-agent --resource openai/api_key --action retrieve