About ReplyAtlas
ReplyAtlas MCP Server lets AI assistants manage ReplyAtlas Instagram operations from MCP clients. It covers comment-to-DM automations, leads, analytics, conversations, and broadcasts, with explicit confirmation required before sending messages to people.
Key features
- 01Create and manage comment-to-DM automations
- 02Find, tag, and export leads
- 03View analytics and DM logs
- 04Review and reply to conversations
- 05Create and manage broadcasts
- 06Require confirmation before sending messages
Use cases
- 01Set up a keyword-triggered DM for Instagram comments
- 02Find and export leads with a chosen tag
- 03Review account analytics and DM activity
- 04Preview and confirm a conversation reply or broadcast
MCP support
Exposes ReplyAtlas account, automation, lead, analytics, conversation, and broadcast tools to MCP clients. Sending a reply or broadcast requires a second call with a confirmation token.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y replyatlas-mcpREPLYATLAS_API_KEYSecretReplyAtlas API key (mf_live_…), created in ReplyAtlas → Settings → API Keys on a plan with API access.REPLYATLAS_BASE_URLOverride the API base URL. Defaults to https://replyatlas.com.Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEReplyAtlas MCP Server
Manage your ReplyAtlas account from Claude, Cursor or any Model Context Protocol client: Instagram comment-to-DM automations, leads, analytics, conversations and broadcasts.
ReplyAtlas auto-sends an Instagram DM to anyone who comments a keyword on your posts, Reels, Lives or story replies, using Instagram's official API. This server lets an AI assistant run that account for you: "create an automation that DMs my free guide to everyone who comments GUIDE", "who were my hottest leads this week?", "export leads tagged webinar".
Requires a ReplyAtlas plan with API access. Create an API key in ReplyAtlas → Settings → API Keys.
Install
No install needed. Your MCP client runs it with npx (Node.js 18+).
Claude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"replyatlas": {
"command": "npx",
"args": ["-y", "replyatlas-mcp"],
"env": { "REPLYATLAS_API_KEY": "mf_live_your_key_here" }
}
}
}
Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"replyatlas": {
"command": "npx",
"args": ["-y", "replyatlas-mcp"],
"env": { "REPLYATLAS_API_KEY": "mf_live_your_key_here" }
}
}
}
Claude Code
claude mcp add replyatlas -e REPLYATLAS_API_KEY=mf_live_your_key_here -- npx -y replyatlas-mcp
Config
| Env var | Required | Default | Notes |
|---|---|---|---|
REPLYATLAS_API_KEY | yes | — | Your mf_live_… API key. |
REPLYATLAS_BASE_URL | no | https://replyatlas.com | Override the API host. |
On start the server calls /api/v1/me to check the key and exits with a clear
message if the key is invalid or the plan has no API access.
Tools
| Area | Tools |
|---|---|
| Account | get_account, list_ig_accounts |
| Automations | list_automations, get_automation, create_automation, update_automation, delete_automation |
| Leads | list_leads, get_lead, list_tags, tag_lead, export_leads |
| Analytics | get_dashboard, get_analytics, list_dm_logs |
| Conversations | list_conversations, reply_to_conversation |
| Broadcasts | list_broadcasts, create_broadcast, send_broadcast |
Safety
Anything that messages real people is two-step. reply_to_conversation and
send_broadcast first return a preview and a confirmToken; nothing is sent
until the tool is called again with that token. The server only talks to the
ReplyAtlas REST API over HTTPS with your API key. It never sees your Instagram
password or tokens.
Development
pnpm install
pnpm test # vitest
pnpm build # tsup → dist/index.js
Links
- Product: https://replyatlas.com
- MCP setup guide: https://replyatlas.com/integrations/mcp
- REST API docs: https://replyatlas.com/docs
License
MIT
Comments
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.