- Segment validated — last 7 days of sign-ups
- Template picked, two subject lines drafted
- A/B test went live with your approval
Your AI agent runs SendNomi on your behalf.
Tell it to build the campaign, pull the report, clean the list — your agent connects to SendNomi over MCP, does the work and reports back. You approve every critical step.
You say it, your agent does it.
Three examples — all three run on real capabilities in the panel today. The agent works through the steps and waits for your approval at the critical point.
- Campaign found, stats read
- Opens, clicks and bounces summarized
- Report dropped into the chat — ready to share
- Recent bounces listed
- Hard failures filtered into a candidate list
- Added to the suppression list with your approval
Your agent’s toolbox: 129 tools.
Four function-based categories. The labels below are representative — the full catalog is published at a single address for developer discovery.
Email operations
From a single email to a full campaign: sending, templates, subject-line testing, address verification.
Contacts & suppression
Add contacts, list with pagination, manage suppression — your list stays clean, permissions stay in place.
Domain & deliverability
DKIM, SPF, DMARC and MX checks; BIMI setup; deliverability reports for any date range.
Analytics & account
Campaign stats, per-message event history, quota and plan usage.
Complex jobs in a single command — your agent asks for the details, gets your approval and delivers the result.
Developer discovery — the full tool catalog:
GET https://api.sendnomi.com/.well-known/mcp/server-card.json
The authority stays with you.
The agent asks for permission, you approve; every call passes scope enforcement and lands in the audit log.
Three steps to your first command.
HTTPS streamable transport — no local runtime, the same structure across all MCP-compatible host applications.
Add the endpoint
One block in your host application’s MCP configuration is all it takes.
{
"mcpServers": {
"sendnomi": {
"url": "https://api.sendnomi.com/mcp"
}
}
} Approve via OAuth
The agent asks for permission on first connection; you pick the scopes on the consent screen.
Give the first command
Type it into the chat — the agent picks the tool, does the work, returns the result.
Prefer an API key instead of OAuth? Generate one in the developer panel with the scopes and validity you choose.
Your first send within minutes.
Create your free account — 500 sends per month, forever.