How do I block a sender from emailing my agent?

Block a sender from emailing your agent with a block rule on the domain. A block match wins over an allow match, whether it hits the envelope sender or the From header.

One sender

POST /domains/{id}/inbound-rules takes kind block and a pattern. The pattern is one address, a bare domain (every address at that domain, not its subdomains), or *@example.com.

curl -sS -X POST https://api.agentisend.com/domains/9c8f8f0e-3d1a-4d3f-9a1e-2b7c1a0f5e42/inbound-rules \
  -H "Authorization: Bearer $AGENTISEND_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{"kind":"allow","pattern":"example"}'

201
{
  "created_at": "2026-09-04T09:14:00.000Z",
  "domain_id": "9c8f8f0e-3d1a-4d3f-9a1e-2b7c1a0f5e42",
  "id": "9c8f8f0e-3d1a-4d3f-9a1e-2b7c1a0f5e42",
  "kind": "allow",
  "note": "example",
  "object": "inbound_rule",
  "pattern": "example"
}

A batch

POST /domains/{id}/inbound-rules/batch/add changes up to 500 patterns in one call. Each item succeeds or fails on its own. GET /domains/{id}/inbound-rules returns the rules. A domain holds 1,000 rules. Past that, DELETE /domains/{id}/inbound-rules/{rule_id} removes one.

curl -sS -X POST https://api.agentisend.com/domains/9c8f8f0e-3d1a-4d3f-9a1e-2b7c1a0f5e42/inbound-rules/batch/add \
  -H "Authorization: Bearer $AGENTISEND_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{"kind":"allow","patterns":["example"]}'

200
{
  "data": []
}

What the refused sender gets

The message is not stored. Nothing is sent back to that sender. GET /emails/receiving/refusals counts those refusals for the domain. The allow list, and the switch between open and allow-list, is How do I limit who can email my AI agent?. This page is the block.

Next

• Email for AI agents
• How do I limit who can email my AI agent?
• How do I give each AI agent its own email address?

Written by the AgentiSend team.