Agent Capabilities

What your AI agent can do — from messaging to browsing to email, all running in a secured, runtime-monitored environment.

Messaging channels

All messaging channels are available on every tier. Connect one or all.

Full two-way messaging with rich media support. DM pairing ensures only approved contacts can interact.

Discord

Channel and DM support with role-aware responses. Your agent joins your server as a bot.

Slack

Workspace integration with channel and direct message support. Thread-aware conversation handling.

Business API integration for customer-facing communication. Contact verification built in.

Browser automation

Your agent can navigate the web, interact with pages, and extract data — all inside a monitored browser environment.

Navigate to URLs and interact with page elements

Fill forms, click buttons, and submit data

Extract structured data from web pages

Take screenshots for verification

Monitored execution — browser activity tracked by runtime security layers

Rate-limited per tier to prevent abuse

Email identity

Pro and Enterprise agents get their own email address for professional communication.

Your agent's own email

Choose a name during setup and your agent gets a dedicated email address. It can send and receive emails, with all HTML content sanitized for security on both inbound and outbound messages.

Inbound emails routed through secure email processing pipeline

Outbound emails sent through dedicated delivery infrastructure

HTML sanitization prevents XSS and content injection

Available on Pro ($159/mo) and Enterprise ($299/mo) plans

AI model access

Your agent has access to 100+ AI models through a multi-model routing layer.

100+ models from leading providers, selected automatically or by preference

Per-agent AI budget included with every tier (overage available via credits)

No conversation data leaves your agent without your explicit configuration

Smart routing optimizes for cost, speed, and capability per request

Tool security

Every tool your agent uses is monitored by our Agent EDR at the kernel level.

Tool execution policy engine evaluates every call before it runs

eBPF kernel probes monitor all process spawns and file access

File integrity monitoring detects unauthorized changes to critical files

Network threat intelligence blocks connections to known-bad IPs

Detection layers run on the host OS, outside the agent's reach

Strict resource limits prevent CPU and memory exhaustion

Tier comparison

All plans include all 7 security layers and all messaging channels.

Feature	Starter$79/mo	Pro$159/mo	Enterprise$299/mo
vCPUs	3	4	8
Memory	4 GB	8 GB	16 GB
Messaging channels	All	All	All
Browser pages	3	5	10
Browser actions / hr	100	500	2,000
Included AI budget	$5 / mo	$15 / mo	$30 / mo
Email identity
Python runtime
Custom skills

All plans include a 5-day free trial, all 7 security layers, and all messaging channels.

Ready to hire your first AI employee?

Production-ready in 5 minutes. All 7 security layers included on every plan.

Get Started See Pricing