mcpField
F
FirstTouch - Allow your agents to operate LinkedIn

FirstTouch - Allow your agents to operate LinkedIn

Remote
firsttouch.comBusiness & Marketing0Updated 8w ago
RepositoryVisit website

About FirstTouch - Allow your agents to operate LinkedIn

A hosted MCP server that connects AI assistants to a FirstTouch workspace for managing audiences, outreach flows, contact enrollment, and tasks. Email and call actions require human approval before execution.

Key features

  1. 01Create audiences from contacts or HubSpot lists
  2. 02Create and manage outreach flow plans
  3. 03Select contacts and track enrollment
  4. 04Review workspace tasks and permissions
  5. 05Require approval for email and call actions

Use cases

  1. 01Build an audience from a connected HubSpot list
  2. 02Draft and publish an outreach flow
  3. 03Enroll selected contacts and check progress
  4. 04Review tasks assigned to workspace members

MCP support

Provides a hosted MCP endpoint at https://mcp.firsttouch.com. MCP clients connect using browser-based OAuth and workspace approval, without entering an API key in the configuration.

#linkedin#sales#outreach#hubspot#crm#prospecting#lead-generation#social-selling#b2b#sales-automation#social-signals

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.firsttouch.ai
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

FirstTouch MCP

Connect your AI assistant to your FirstTouch workspace to build audiences, create outreach flows, start selected contacts, track enrollment progress, and review tasks in natural language.

Connect

Use the hosted FirstTouch MCP server:

https://mcp.firsttouch.com

The client opens a browser for FirstTouch sign-in and workspace approval. Authentication uses OAuth, so no API key needs to be pasted into the MCP configuration.

Generic MCP client configuration:

{
  "mcpServers": {
    "firsttouch": {
      "type": "http",
      "url": "https://mcp.firsttouch.com"
    }
  }
}

What you can do

  • Check the signed-in user, active workspace, team members, and assignment permissions.
  • List HubSpot contact lists connected to the workspace.
  • Create audiences from supplied contacts or a connected HubSpot list.
  • Add contacts to an audience and attach an audience to a flow plan.
  • List, inspect, and create outreach flow plans with email, call, manual, wait, and branch actions.
  • Review available flow variables and publish or unpublish a valid flow.
  • Inspect awaiting contacts, start only the selected contacts, and follow enrollment progress.
  • Review tasks assigned to workspace members.

Every email and call outreach action created through this connection requires human approval before execution.

Example requests

  • "Show my FirstTouch user, active workspace, and assignable team members."
  • "Create a FirstTouch audience from my HubSpot list named Q3 Enterprise Leads."
  • "Draft a flow that sends an intro email, waits one day, and creates a call task."
  • "Show the awaiting contacts for this flow and enroll the first five."
  • "List my FirstTouch tasks."

Network access and data

The plugin connects to the FirstTouch-hosted MCP endpoint at https://mcp.firsttouch.com. Browser-based authorization is handled by FirstTouch services.

FirstTouch accesses records from the workspace selected during authorization and respects that user's permissions. When a user explicitly requests a workflow involving a connected HubSpot account, FirstTouch may read the relevant HubSpot data. The plugin does not install or execute local code.

Links

License

Licensed under the Apache License 2.0.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.