instantly-mcp
MCP server for Instantly.ai, allowing Claude to query campaign analytics, warmup health, and sending accounts, and to manage campaigns and replies with confirmation.
README
instantly-mcp
A local stdio MCP server that exposes the Instantly.ai v2 API to Claude Code — the same way crmdata exposes HubSpot, but with no dashboard, no database, no hosting. Just the tools.
Claude Code ──stdio──▶ node src/server.mjs ──HTTPS+Bearer──▶ Instantly v2 API
Setup
cd instantly-mcp
npm install
cp .env.example .env # then paste your INSTANTLY_API_KEY
npm run smoke # boots the server, lists tools, does one live read
Get the key from Instantly → Settings → Integrations → API keys.
Wire it into Claude Code
Use absolute paths (stdio launches the process directly):
claude mcp add instantly -- node --env-file="C:/Users/colen/OneDrive/Desktop/Code-Projects/instantly-mcp/.env" "C:/Users/colen/OneDrive/Desktop/Code-Projects/instantly-mcp/src/server.mjs"
Verify:
claude mcp list
Then just ask, in any Claude Code session:
"How are my Instantly campaigns doing this week, and is any sending account's warmup health dropping?"
Tools
Read
| Tool | What it does |
|---|---|
list_campaigns |
Campaigns with ids, names, status |
get_campaign_analytics |
Sent/open/reply stats (one campaign or all) |
list_accounts |
Connected sending inboxes |
warmup_analytics |
Inbox-vs-spam + 0–100 health per account |
account_vitals |
SPF/DKIM/DMARC deliverability checks |
list_emails |
Unibox (address or thread:<id>) |
get_email |
One email, full body |
unread_count |
Unread count |
list_leads_in_campaign |
Leads on a campaign |
Write — staging (no send)
| Tool | What it does |
|---|---|
create_campaign |
Create a sequence + schedule (does not launch) |
add_leads |
Bulk-add leads to a campaign |
mark_thread_read |
Mark a unibox thread read |
Write — LIVE SEND (require confirm: true)
| Tool | What it does |
|---|---|
activate_campaign |
Launches a campaign — it starts sending |
reply_to_email |
Sends a reply into a unibox thread |
The two live-send tools refuse unless you pass confirm: true, so cold email is never fired on a one-word ask.
Files
src/client.mjs thin Instantly v2 fetch wrapper (zero deps)
src/tools.mjs tool definitions (name, description, zod schema, handler)
src/server.mjs McpServer + stdio transport
scripts/smoke.mjs end-to-end boot + tools/list test
Notes
- All 9 read tools are verified live against the real workspace (2026-07-17). Two endpoints ported from salesintel were stale against the current v2 API and were corrected here: warmup is
POST /accounts/warmup-analytics(hyphen), and vitals takes{ accounts: [...] }and returns per-domain MX/SPF/DKIM/DMARC.client.mjstags each endpoint[verified]/[ported]. - The 5 write tools (
create_campaign,add_leads,mark_thread_read,activate_campaign,reply_to_email) are not exercised by the smoke test — they create/send.list_leads_in_campaignshares the/leads/listpath withadd_leadsand is verified, but confirm the add shape on your first real use. - This repo does the querying; the
salesintelapp still owns durable infra (webhooks, reply polling, warmup ramp). Two views on one Instantly account, not two sources of truth.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.