canvassed developer docs

Connect the canvassed MCP

The canvassed MCP lets you ask an AI assistant about commercial real estate listings in plain language. You can connect it to a supported AI app and sign in when it asks. You don’t enter an API key in the AI app.

To connect, sign in with a canvassed account that includes AI assistant access. Start a free 14-day trial, or add AI assistant access to Pro. You'll go straight back to your AI app once you're set up.

canvassed listings come from publicly available brokerage listing pages, and every listing links back to its source. Brokers can request an update or removal at info@canvassed.ai. Data sources and removal

Before connecting, review how AI assistant connections use and log data.

Developer docs homeGet help
canvassed MCP server URLhttps://mcp.canvassed.ai/mcp

Overview

The canvassed MCP is a remote connector that lets an AI assistant find and review commercial real estate listings. After you connect, the assistant can discover the available tools and choose one or more based on what you ask. You can describe what you need in plain language; you don’t need to know a tool’s name or enter its parameters, and you don’t use an API key to sign in to the MCP.

Some tools return listings, counts, or parcel details directly. Others are helper tools that resolve a place, company, or filter value. The AI app receives the tool results and presents them in the conversation, so the exact wording and layout can vary by platform.

What is an MCP server?

An MCP server connects an AI app to a service by providing a standard way to discover and use its tools. You can read the official introduction to MCP ↗ for a short overview of the protocol.

Prerequisites

  • You’ll need an AI app and plan that supports adding a remote MCP connector. Some workspaces require an administrator to enable custom connectors.

Connect to your platform

You can choose one of these AI apps. Select its name to open the setup steps.

Claude

Custom remote connectors are available in Claude on supported Pro, Max, Team, and Enterprise plans. A Team or Enterprise owner may need to allow the connector first.

  1. In Claude, open Settings → Connectors.
  2. Select Add custom connector.
  3. Enter the canvassed MCP server URL: https://mcp.canvassed.ai/mcp.
  4. Select Add, then select Connect for canvassed.
  5. Sign in to your canvassed account and approve the connection.
  6. In a Claude conversation, open Search and tools and enable canvassed.

Claude’s current custom connector instructions ↗

ChatGPT

ChatGPT connects to remote MCP servers through custom apps. Availability depends on your ChatGPT plan and workspace settings; a workspace administrator may need to enable developer mode or create and publish the app.

  1. In ChatGPT, open Settings → Apps and enable developer mode if it’s available for your account.
  2. Select Create app or Create in the Apps area.
  3. Enter the canvassed MCP server URL: https://mcp.canvassed.ai/mcp.
  4. Choose OAuth authentication, then scan the tools.
  5. When prompted, sign in to your canvassed account and approve the connection.
  6. Save the app. If your workspace requires review, ask an administrator to publish or enable it.
  7. Start a chat, open the tools menu, and select the canvassed app.

ChatGPT’s current MCP app instructions ↗

Perplexity

Perplexity supports custom remote MCP connectors on supported plans. Your account or organization may control whether you can add one.

  1. In Perplexity, open Account settings → Connectors.
  2. Select Custom connector, then choose Remote.
  3. Enter a name such as canvassed and the server URL https://mcp.canvassed.ai/mcp.
  4. Choose OAuth authentication and follow the prompts.
  5. Sign in to canvassed and approve access.
  6. Enable the connector when you ask a question in Perplexity.

Perplexity’s remote MCP setup instructions ↗

Gemini

Gemini supports custom MCP apps in the Gemini web app. Google currently limits this feature to personal Google Accounts in the US, with Keep Activity turned on; it’s not available for work or school accounts.

  1. In a desktop browser, open gemini.google.com.
  2. Open Settings → Connected Apps. If you don’t see it, open Personal Intelligence first.
  3. Under Custom apps, select Add a custom app.
  4. Enter the canvassed MCP server URL: https://mcp.canvassed.ai/mcp, then select Next.
  5. Follow the sign-in prompts and approve the canvassed connection.
  6. In a prompt, type @ and choose the canvassed app.

Google’s current custom app instructions ↗

Gemini Enterprise has a separate administrator setup and may require OAuth configuration by your organization.

Try a question

After you connect and enable canvassed, you could ask:

“Find retail listings for lease in Austin, Texas.”

This asks canvassed to make a real request using the connection you approved. The result may use your account’s MCP allowance.

Troubleshooting

  • The connector option isn’t available: your plan or workspace settings may not support custom MCP connectors. Ask your administrator or check your AI app’s current requirements.
  • canvassed says your account can’t connect: review the access options above, then try connecting again.
  • Connection succeeds but no tools appear: refresh the connector in the AI app, then start a new conversation and enable canvassed.

If you need help with setup, you can contact canvassed.

Developer docs homeNext: MCP tools and help