Build with AI coding agents

Your coding agent can build on ReachHunt for you. Give it two MCP servers: one to work in your workspace, and one to search these docs. Every page is also available as plain markdown, so the agent reads the same docs you do.

What your agent gets

MCP server	URL	What it does	Sign-in
ReachHunt	`https://mcp.reachhunt.ai/mcp`	Reads and acts in your workspace: inbox, messages, lead lists, campaigns and agents	ReachHunt sign-in
ReachHunt docs	`https://docs.reachhunt.ai/mcp`	Searches these docs and reads full pages	None

Both servers use Streamable HTTP over HTTPS. You don’t run anything locally.

Connect your agent

Run these in your terminal:

claude mcp add --transport http reachhunt https://mcp.reachhunt.ai/mcp
claude mcp add --transport http reachhunt-docs https://docs.reachhunt.ai/mcp

Start Claude Code, type /mcp, choose reachhunt and select Authenticate. A ReachHunt page opens in your browser. Choose your workspace and select Connect.

Paste this into ~/.cursor/mcp.json, or .cursor/mcp.json inside one project:

mcp.json

{
  "mcpServers": {
    "reachhunt": {
      "url": "https://mcp.reachhunt.ai/mcp"
    },
    "reachhunt-docs": {
      "url": "https://docs.reachhunt.ai/mcp"
    }
  }
}

In Cursor Settings, open MCP and select the sign-in button next to reachhunt.

Paste this into .vscode/mcp.json in your project:

.vscode/mcp.json

{
  "servers": {
    "reachhunt": {
      "type": "http",
      "url": "https://mcp.reachhunt.ai/mcp"
    },
    "reachhunt-docs": {
      "type": "http",
      "url": "https://docs.reachhunt.ai/mcp"
    }
  }
}

Select Start above reachhunt in the file, then sign in when VS Code asks.

Run these in your terminal:

codex mcp add reachhunt --url https://mcp.reachhunt.ai/mcp
codex mcp add reachhunt-docs --url https://docs.reachhunt.ai/mcp
codex mcp login reachhunt

Run these in your terminal:

gemini mcp add --transport http reachhunt https://mcp.reachhunt.ai/mcp
gemini mcp add --transport http reachhunt-docs https://docs.reachhunt.ai/mcp

Start Gemini CLI and type /mcp auth reachhunt to sign in.

Paste this into ~/.codeium/windsurf/mcp_config.json:

mcp_config.json

{
  "mcpServers": {
    "reachhunt": {
      "serverUrl": "https://mcp.reachhunt.ai/mcp"
    },
    "reachhunt-docs": {
      "serverUrl": "https://docs.reachhunt.ai/mcp"
    }
  }
}

For Claude, ChatGPT and other chat apps, follow MCP server.

Docs built for agents

Resource	What it gives your agent
`llms.txt`	An index of every page, with a one-line summary and link
`llms-full.txt`	Every page in one file, ready to paste into a prompt or index
`.md` on any page	Add `.md` to a page URL to get its markdown, for example `/developer-api/authentication.md`
Page menu	Open the menu at the top of any page to copy it as markdown, or open it in ChatGPT, Claude, Perplexity, Cursor or VS Code

Try this prompt

Paste this into your agent after you connect both servers:

Using the ReachHunt docs MCP, build a small Node.js server that receives
ReachHunt webhooks. When a campaign.lead_replied event arrives, check the
X-Webhook-Secret header, then post the lead's name and reply to Slack.

The agent looks up the event payload and the webhook guide in the docs, writes the handler, and checks the secret header with a constant-time comparison.

Tell your agent

These are the details agents most often get wrong:

Every response has the same envelope. Read data when status is true, and error when it is false. See Response format.
Items carry a platform. It is x or linkedin. Many fields differ between the two.
Sending limits apply per account. The app, campaigns and the API share them. Check them with Get account limits before you send in bulk. See Rate limits.
ReachHunt doesn’t sign webhooks. Add a secret header and compare it in constant time. See Receive webhooks.

The ReachHunt MCP server acts as the workspace owner and sends real messages from your connected accounts. Try new agent code on a test workspace first, and never commit an API key.

Next steps

MCP server

What the ReachHunt MCP server can do, and its limits.

Quickstart

Make your first API request.

Webhooks

Get events sent to your server as they happen.

Getting started

Webhooks

Reference

Event types

API reference

Build with AI coding agents

What your agent gets

Connect your agent

Docs built for agents

Try this prompt

Tell your agent

Next steps

MCP server

Quickstart

Webhooks

​What your agent gets

​Connect your agent

​Docs built for agents

​Try this prompt

​Tell your agent

​Next steps

MCP server

Quickstart

Webhooks

What your agent gets

Connect your agent

Docs built for agents

Try this prompt

Tell your agent

Next steps