Articles on: SalesBlink AI

Connect SalesBlink MCP to Your AI Tool

title: Connect SalesBlink MCP to Your AI Tool
description: Control SalesBlink from any MCP-compatible AI assistant.

Overview


Use the Model Context Protocol (MCP) to control SalesBlink directly from your AI assistant. Create lists, launch sequences, check replies, track reports and many more — all through natural language commands.


Prerequisites


Before you start, confirm you have the following:


  • Active SalesBlink account: Sign up at https://run.salesblink.io/signup if you do not have one.
  • SalesBlink API key: Retrieve it from SalesBlink Dashboard > Integrations > API.
  • MCP-compatible AI client: Cursor, Claude Desktop, VS Code, Windsurf, Zed, or any client listed below.


Step-by-Step Instructions


Step 1: Copy Your API Key


Open your browser and navigate to your SalesBlink account.


  • Click Account in the Top Navbar.


  • Select Integrations from the dropdown.


  • Click the API tab.


  • Click Create API Key


  • Enter a name for your API key (e.g., "Claude Connect ")
  • Click Create New API Key.
  • Copy API Key







Step 2: Open the MCP Configuration Page


Go to https://mcp.salesblink.io in your browser.


  • Locate the MCP Client Setup Guide section.


  • Paste your API key into the Enter your SalesBlink API Key field.


  • Click Apply. The page automatically inserts your key into all configuration snippets below.






Step 3: Select Your AI Client


Scroll to the client list and find your AI tool. Click the card to expand setup instructions.




Step 4: Configure Your Client


Follow the exact steps for your tool.


Cursor (AI Code Editor)


  • File Method (Recommended): Open or create ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-level).


  • Paste the Core Configuration Block from the MCP page.


  • Save the file and restart Cursor.



Windsurf (AI Code Editor by Codeium)


  • Open or create ~/.codeium/windsurf/mcp_config.json.


  • Add a remote server using serverUrl.


  • Restart Windsurf.


  • Click the MCPs icon (hammer) in the Cascade panel to verify it is enabled.


{ "mcpServers": { "salesblink": { "serverUrl": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY" } } }


Zed (High-Performance Editor)


Note: Zed uses context_servers instead of mcpServers.


  • Open Zed settings (Cmd/Ctrl + ,) or edit ~/.config/zed/settings.json.


  • Add this block.


  • Save the file and restart Zed.


"context_servers": { "salesblink": { "enabled": true, "url": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY" } }


Claude Desktop (Official Anthropic Client)


Note: Claude Desktop remote MCP uses Connectors (paid tiers). If Connectors are unavailable, use the Node.js fallback below with npx mcp-remote. Install Node.js first if needed.



  • Mac: ~/Library/Application Support/Claude/claude_desktop_config.json


  • Windows: %APPDATA%\Claude\claude_desktop_config.json


  • Use the fallback block below in that file.


  • Save and restart Claude Desktop.


{ "mcpServers": { "salesblink": { "command": "npx", "args": [ "-y", "mcp-remote", "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY", "--transport", "streamable-http" ] } } }


Codex (by OpenAI)


  • Method 1 - CLI Command (Fastest): Run the command below.


codex mcp add salesblink --url https://mcp.salesblink.io/mcp?api_key=YOUR_KEY


  • Method 2 - Configuration File (config.toml): Open your Codex configuration file:


  • Global: ~/.codex/config.toml


  • Project: .codex/config.toml


  • Add the following block and save.


[mcp_servers.salesblink] url = "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY"


Claude Code (CLI) - Anthropic's CLI Tool


  • Run the CLI command below.


  • Restart your terminal or run /mcp inside Claude Code to verify.


claude mcp add --transport http salesblink https://mcp.salesblink.io/mcp?api_key=YOUR_KEY


VS Code (Cline / Roo Code)


  • Open the Cline or Roo Code extension panel in VS Code.


  • Click the MCP Servers icon (plug/server icon at the bottom).


  • This opens cline_mcp_settings.json or roo_mcp_settings.json.


  • Add a remote entry with url and type: "streamableHttp", then save.


  • A green indicator will light up when connected.


{ "mcpServers": { "salesblink": { "url": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY", "type": "streamableHttp" } } }


GitHub Copilot (VS Code Native Extension)


  • Open VS Code settings.json (Command Palette > Open User Settings JSON).


  • Add the Copilot-specific configuration:


  • Save the file and restart VS Code.


"github.copilot.mcp.servers": { "salesblink": { "type": "http", "url": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY", "tools": ["*"] } }


JetBrains IDEs (IntelliJ, PyCharm, etc.)


  • Go to Settings > Tools > AI Assistant > MCP Servers.


  • Choose HTTP.


  • Add the JSON configuration.


{ "mcpServers": { "salesblink": { "url": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY" } } }


OpenHands (AI Software Development Agent)


  • OpenHands supports remote HTTP via CLI. Run the command below.


  • Restart your OpenHands session if needed.


openhands mcp add salesblink --transport http https://mcp.salesblink.io/mcp?api_key=YOUR_KEY


Kimi Code & Kilo Code (AI Coding Assistants)


  • Kimi Code: Open ~/.kimi/mcp.json or click MCP Servers in the VS Code extension gear menu. Add a remote server using url.


{ "mcpServers": { "salesblink": { "url": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY" } } }


  • Kilo Code: Go to VS Code pane > Settings > Agent Behaviour > MCP Servers. Edit Global MCP (mcp_settings.json). Add a remote MCP entry and save.


{ "mcp": { "salesblink": { "type": "remote", "url": "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY", "enabled": true } } }


Other MCP Clients (Custom / CLI Clients)


For any other client conforming to the standard MCP specification:


  • Look for an mcp.json or config.json file in the tool's root directory or ~/.config/ folder.


  • Paste the standard mcpServers object.


  • If the client does not support remote URL directly, use Node.js + mcp-remote:


  • Need Node.js? Download (LTS) or see the step-by-step setup guide.


{ "mcpServers": { "salesblink": { "command": "npx", "args": [ "-y", "mcp-remote", "https://mcp.salesblink.io/mcp?api_key=YOUR_KEY", "--transport", "streamable-http" ] } } }




Step 5: Verify the Connection


Test that your AI tool can reach SalesBlink.


  • Open your AI client's chat panel.


  • Type: "Show me my SalesBlink lists"or"What sequences are running?"


  • The AI should respond with live data from your SalesBlink account.




Tip: If you get an error, check that your API key is pasted correctly and that the URL matches exactly what is shown on the MCP page.


What You Can Do


Once connected, use natural language commands in your AI tool:


  • Create Lists: Build and manage prospect lists directly from your AI assistant.

Example: "Create a new list called SaaS Founders"


  • Launch Sequences: Start personalized email sequences with a simple command.

Example: "Launch my Enterprise Outreach sequence to the SaaS Founders list"


  • Check Inboxes: Monitor replies and manage conversations from your AI tool.

Example: "Show me unread replies from today"


  • Track Reports: Get real-time analytics and campaign performance insights.

Example: "What was my open rate last week?"


  • Run Inbox Placement Tests: Ensure your emails land in the inbox, not spam.

Example: "Run an inbox placement test for sender noreply@company.com"


  • Verify Emails: Verify email addresses before adding them to your list.

Example: "Verify these email addresses before adding them to my list"


Example Commands


Here are more examples of what you can ask your AI assistant:


Task

Example Command

Create a list

"Create a list called Marketing Agencies in New York"

Add prospects

"Add these emails to my SaaS Founders list"

Launch a sequence

"Launch the Welcome Sequence to my new list"

Check replies

"Show me replies from this week"

View analytics

"How many opens did my campaign get yesterday?"

Run a test

"Run an inbox placement test for outreach@mycompany.com"

Verify emails

"Verify john@example.com and sarah@example.com"


Troubleshooting


Connection Failed


  • Confirm your API key is active in SalesBlink Dashboard> Integrations> API.



  • Restart your AI client after any configuration change.


AI Does Not Recognize Commands


  • Verify the MCP server shows as connected (green indicator) in your client.


  • Try simpler commands first, like "List my campaigns".


  • Check that the SalesBlink MCP server is selected as an active tool in your AI chat interface.


Rate Limit Errors


  • The SalesBlink API limits: 30 GET requests/minute, 15 POST/PATCH requests/minute, 10 PUT/DELETE requests/minute.


  • Wait 60 seconds and retry. Contact SalesBlink support if you need higher limits.


Frequently Asked Questions


Which AI tools are compatible with SalesBlink MCP?
SalesBlink MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (with Cline/Roo Code), Windsurf, Zed, GitHub Copilot, Codex, Kimi Code, Kilo Code, and many more.


Do I need coding skills to use SalesBlink MCP?
Not at all. You interact with it using plain English. Simply tell your AI assistant what you want to do, and the AI handles the technical details for you.


Is SalesBlink MCP secure?
Yes. SalesBlink MCP uses enterprise-grade authentication with API keys. Your credentials are never stored by the MCP client. All communications are encrypted, and you can revoke access at any time from your SalesBlink dashboard.

Updated on: 03/07/2026

Was this article helpful?

Share your feedback

Cancel

Thank you!