Provisioned inboxes
Email and SMS addresses on your domain, ready in one API call. No MX record wrangling.
Inbox infrastructure for AI agents
Get your agent an email and a phone number
Set up secure inboxes in seconds, buy phone numbers, block spam before it reaches your agent, and route clean messages anywhere you need them.
Shield
Webhooks
Storage
Routing
Email
SMS
Live agent message stack. Provision, guard, route, deliver.
claude — inbounter mcp
$ claude --mcp inbounter
> "Give my agent an email address and a phone number"
⚡ inbounter › create_inbox
name "shopping-agent"
provisionSource true
shieldEnabled true
✓ email shopping-agent@inbounter.com
✓ phone +1 (415) 555-0147
✓ shield enabled
Your agent gets a real email address and phone number in one API call. Works with Claude Code, Cursor, OpenClaw, and any agent that supports MCP or HTTP.
Get your agent a free email now
Claim your free agent email address
@
Developer API
Set up in under five minutes.
Install the SDK, create an inbox, and start receiving messages. No configuration files, no SMTP settings, no DNS headaches.
Step 2Run the OpenClaw identity setup command
openclaw run setup-agent-identityWorks with
$ openclaw run setup-agent-identity
? Inbox provider Inbounter
? Agent name shopping-agent
? Inbound webhook https://agent.app/inbound
? Sync Gmail too Yes
✓ secure email shopping-agent@inbounter.com
✓ secure phone +1 (415) 555-0147
✓ shield mode enabled
✓ gmail sync readyWhat you get
Everything agents needs to send and receive messages.
Secure inboxes, phone numbers, spam protection, and message routing — all ready in minutes, not days.
Shield Mode
One flag to enforce allow/reject sender rules and quarantine suspicious payloads before they touch your agent.
Structured routing
Route to webhooks, Postgres, Supabase, Slack, or any HTTP endpoint. Retries and delivery logs included.
Message storage
Quarantined and processed messages stored with full metadata. Semantic search across your inbox history.
Agent-first API
Node.js and Python SDKs with TypeScript types. REST API for every runtime that speaks HTTP.
OpenClaw ready
Native integration with OpenClaw flows. Works with any agent framework that accepts structured inbound events.
Features
Your agent communications secured and taken to the next level.
Stop spam, lock down send permissions, and stream live event updates so your team can operate agent messaging with confidence.
ALLOW @openai.comALLOW @anthropic.comREJECT phishing.ruVERIFY billing@acme.ioREJECT spoofed domain
Protect your agent from spam
Filter spoofed domains, suspicious senders, and noisy traffic before messages reach your runtime.
Secure sandbox and permissions
Restrict which addresses your agent can message and enforce policy gates before outbound actions execute.
EMAIL_RECEIVEDSMS_RECEIVEDSPAM_REJECTEDWEBHOOK_SENT
Get updates in real time
See inbound events as a live activity tray and stream them to your webhooks the moment they happen.
700+ Apps
SlackGitHubHubSpotOpenClawPlug into 700+ integrations
Route events into CRMs, support tools, data warehouses, and ops stacks without writing glue code.
Delivery +18.2%
See delivery performance instantly
Track throughput, retries, and delivery quality by inbox and channel in one clear timeline.
shield.passwebhook.200retry.queued
Spike detected+12% engagementanomaly routed
Catch anomalies before incidents escalate
Surface unusual traffic and engagement spikes in real time, then trigger automated response workflows.
How it works
From zero to live in three steps.
1
Create an inbox
One API call provisions an email address on your domain with default routing configured.
2
Enable Shield
Set allow and reject sender rules. Suspicious messages get quarantined before your agent runs.
3
Route anywhere
Clean messages are forwarded as structured payloads to your webhook, database, or automation.
Start building today.
Free to start. No credit card required.
Viewing as