Control WarmySender from your AI assistant

Your Claude, Cursor, Windsurf — now connected to your cold email campaigns, email warmup, LinkedIn outreach, and multichannel flows. One install, full control, safety limits enforced.

What this is

WarmySender covers all four pillars of modern outreach: cold emailing, email warmup, LinkedIn outreach, and multichannel sequences. The /ai-assistants page is the public install hub for connecting AI clients to WarmySender via the Model Context Protocol (MCP).

One install gets your AI assistant full natural-language control of your workspace. Your AI can list and create cold email campaigns, enroll prospects, check email and LinkedIn account health, pause and resume campaigns, update warmup settings, and manage suppression lists — all through the same safety-checked scheduler that the WarmySender web UI uses. Looking for the deep technical docs? See the full MCP integration guide.

Supported AI clients

Five clients ship with one-paste install snippets. Any other MCP-compatible client (Continue, Cline, Goose, custom agents) also works — point it at https://warmysender.com/mcp with your Bearer token.

Claude Desktop

Native desktop app from Anthropic. Paste one snippet into claude_desktop_config.json and restart — your assistant runs cold email, warmup, and LinkedIn campaigns from chat.

Claude Code

Anthropic's CLI for software engineering. One claude mcp add command registers WarmySender — perfect for engineers who want to manage outreach from the terminal.

Cursor

AI-first code editor. Drop a .cursor/mcp.json into your project root — Cursor speaks streamable-http natively, no launcher needed.

Windsurf

Codeium's agentic IDE. Add WarmySender to ~/.codeium/windsurf/mcp_config.json and Cascade can drive your campaigns alongside your code.

Zed

High-performance multiplayer code editor. Add WarmySender under context_servers in settings.json — your AI panel can now run outreach without leaving the editor.

What your AI can do

30 tools, every one scope-gated and workspace-scoped. Grouped by impact:

Read tools (14)

Query workspace state without side effects. Safe for the AI to call freely. Workspace-scoped — your key only sees data from its own workspace.

Write tools (13)

Modify campaigns, prospects, and suppressions. Destructive actions (pause, unenroll, suppress) carry a destructiveHint so the AI prompts you for confirmation before running them.

LinkedIn tools (3)

Safety-gated LinkedIn actions. Every call routes through the scheduler — daily, weekly, and 4-week ramp limits always win. The AI cannot connect or disconnect accounts.

Safety-gated. Routed through the scheduler — daily, weekly, and ramp limits always apply.

Safety is always on

LinkedIn daily and weekly limits, warmup ramps, and domain hard-caps are enforced by the scheduler regardless of what your AI asks for. Your AI cannot connect or disconnect accounts.

Five guardrails always apply:

  1. Workspace scoping on every call. Your API key resolves to exactly one workspace. There is no path to read or mutate another tenant's data.
  2. Scheduler-first execution. LinkedIn invite, message, and view calls never happen inline from MCP. The scheduler owns the rate limits, ramp schedule, and per-account safety caps. Asking the AI to enroll 1,000 prospects still sends at the safe rate.
  3. LinkedIn daily/weekly limits and the 4-week ramp are non-negotiable. A new account starts at ~10 invites/day and grows over four weeks to safe maximums (50 invites/day, 150 messages/day). MCP cannot raise these — account safety always wins over speed.
  4. Per-mailbox sending pacer and warmup [ref: XXXX] identifier stay intact. No tool exposed to the AI modifies or removes the warmup subject-line tag or bypasses per-mailbox daily caps.
  5. No connect / disconnect / delete tools. Connecting and disconnecting email or LinkedIn accounts stays in the UI. There are no delete_* tools — soft-cancel via pause and archive only.

More on LinkedIn safety: LinkedIn Safety FAQ.

Pricing & tiers

MCP usage is included in every paid plan — no extra add-on charge. Free-plan users see Upgrade to connect on this page; Starter and above can install.

Plan Monthly price MCP calls / month
Starter (legacy) $6.99 1,000
Pro $14.99 10,000
Business $29.99 100,000
Enterprise / Agency $69.99+ Unlimited

Per-minute rate limits also apply (120 reads/min, 30 writes/min, 10 LinkedIn writes/min, 5 bulk/min — per workspace). LinkedIn add-on seats ($9/seat/month) are still required if you call LinkedIn tools. See the pricing page for the full breakdown and annual billing options.

Quick start in 3 steps

  1. Create an API key. Go to Settings → API Keys and pick the scopes you want to expose. Save the key value — you only see it once.
  2. Paste the install snippet into your client. Pick Claude Desktop, Claude Code, Cursor, Windsurf, or Zed. Replace the API-key placeholder with your real key. Full snippets and copy-paste configs live on the MCP server docs page.
  3. Ask your assistant to use WarmySender. Try “list my campaigns” or “what's my LinkedIn account health?” The client discovers MCP tools automatically.

Frequently asked questions

Which AI assistants can I connect?

Claude Desktop, Claude Code, Cursor, Windsurf, and Zed are officially supported with one-paste install snippets. Any other Model Context Protocol (MCP) client that speaks Streamable HTTP also works — point it at https://warmysender.com/mcp with a Bearer token.

What can my AI actually do once connected?

Drive all four pillars of WarmySender: cold email campaigns, email warmup, LinkedIn outreach, and multichannel sequences. 30 tools total — list and create campaigns, enroll prospects, check mailbox and LinkedIn health, pause or resume campaigns, manage suppression lists, and update warmup settings. Connecting and disconnecting accounts stays in the UI for safety.

Can my AI accidentally burn my sending reputation or get my LinkedIn banned?

No. Every LinkedIn action routes through our scheduler, which enforces Unipile/LinkedIn per-account daily and weekly limits plus the 4-week ramp schedule. Email sends respect the per-mailbox pacer, domain caps, and warmup ramp. If your AI asks to enroll 1,000 prospects at once, the scheduler still sends at the safe rate. MCP cannot bypass any of these limits — account safety always wins.

Who can use the AI Assistants integration?

Free plan users see "Upgrade to connect". Starter and above (Starter $6.99/mo legacy, Pro $14.99/mo, Business $29.99/mo, Enterprise $69.99/mo) can install. MCP usage is included in the subscription — no extra add-on charge.

How many calls do I get per month?

Starter includes 1,000 MCP calls per month, Pro includes 10,000, Business includes 100,000, and Enterprise/Agency tiers are unlimited. Per-minute rate limits apply on top: 120 reads/min, 30 writes/min, 10 LinkedIn writes/min, 5 bulk/min — per workspace.

How do I revoke access?

Go to Settings > API Keys and revoke the key. Revocation is instant — the next call from any AI client using that key gets an unauthorized error. You can also narrow scopes on a key to tighten what the AI is allowed to do without fully revoking.