How to Connect Cursor to Salesforce

What You Need

Connect Cursor to Salesforce Through an MCP Server

1. 1

   Add the server in an mcp.json file. Cursor's MCP documentation says a project file at .cursor/mcp.json holds tools for that project, and a file at ~/.cursor/mcp.json holds tools available in every project. A remote server uses a url field. Paste the URL your Salesforce MCP server gives you.

{ "mcpServers": { "salesforce": { "url": "<your-mcp-server-url>" } } }

2. 2

   Open Customize in the sidebar and confirm the server is listed. The same documentation says you install and manage MCP servers from Customize, and you enable or disable them there.

3. 3

   Complete the authorization if Cursor opens one. With Cirra AI, the screen is the Cirra AI MCP Server authorization request. It lists openid, email, profile and offline_access. You already approved Salesforce access when you added the org in your Cirra AI account.

4. 4

   Check how Cursor approves tool calls before you rely on the connection. Cursor's run modes page says to open Settings > Agents > Approvals & Execution. Auto-review lets allowlisted MCP tools run immediately and sends other calls to a classifier. Run Everything runs every call. Allowlist with an empty allowlist behaves like the retired Ask Every Time mode.

5. 5

   Start a new chat and describe a Salesforce task to confirm the connection is live.

Verify the Connection

Ask Cursor something specific to your org, such as “What custom objects exist in my Salesforce org?” If it returns your org’s actual custom objects, the connection is working. If it gives a generic answer, the connection did not initialize. Re-check the URL in mcp.json, confirm the server is enabled under Customize in the sidebar, and restart Cursor.

What Cursor Can Do Once Connected

This depends entirely on which Salesforce MCP server you connect, not on Cursor itself. An operational-only server lets Cursor read and update records and run SOQL queries. A server with metadata write access lets Cursor also create objects and fields, build Flows, and manage permission sets, directly from chat, alongside your regular coding work.

Troubleshooting

Server doesn’t appear as connected. Check the JSON syntax in mcp.json for a missing comma or bracket, confirm the server is enabled under Customize in the sidebar, then restart Cursor.

Cursor gives generic answers instead of org-specific ones. Confirm the server is enabled under Customize in the sidebar and start a new chat.

Authorization fails during setup. Confirm you’re authorizing with a Salesforce user that has the permissions the connector needs.

Frequently Asked Questions

Using Cirra AI as Your Salesforce MCP Server

Cirra AI is the Salesforce MCP server used in the example above. Unlike Salesforce’s own hosted MCP server, the Cirra AI Salesforce Admin MCP Server writes to the metadata layer directly in conversation: creating objects and fields, building Flows, and managing permission sets and page layouts.

Governance is built into every connection: granular, category-level access controls plus an exportable audit trail mean you stay in control of what Cirra AI can touch. If your team manages more than one Salesforce org, a multi-team hierarchy lets you connect and switch between them from a single setup, rather than reconfiguring per project.

A free plan is available with no credit card required. Connecting your first org takes under 10 minutes.

For the full picture of what’s available across Salesforce MCP servers, see the Salesforce MCP Server guide. If your team uses Claude instead, see connecting Claude to Salesforce.