Build, deploy, and operate AI agents on Bothive.
These docs cover the pieces behind production agents: HiveLang, memory, integrations, tool execution, channels, schedules, the Execute API, and observability.
Start here
Choose the path closest to what you are building.
Quick Start
Create your first Bothive agent, run it locally, and deploy it to the hosted runtime.
HiveLang
Learn the language for defining bots, tools, memory, events, schedules, and production behavior.
Core Concepts
Understand agents, swarms, memory, tools, triggers, proactive jobs, and workflow execution.
Integrations
Connect Slack, Notion, GitHub, email, APIs, databases, and the tools your team already uses.
Execute API
Call Bothive agents from your backend, app, internal tooling, or customer-facing product.
Guides
Follow practical paths for embedding bots, connecting apps, and turning workflows into agents.
Agent contract
Describe the behavior. Bothive runs the system around it.
HiveLang keeps agent behavior readable while the hosted runtime handles model calls, tools, memory, channels, approvals, schedules, and traces.
- Define agent behavior in the builder or HiveLang.
- Attach memory, tools, integrations, and approval rules.
- Deploy to web, API, schedules, Slack, WhatsApp, or your own app.
- Inspect traces, tool calls, usage, and failures from the dashboard.
agent SupportAgent {
model fast
memory scoped by user.id
tools {
knowledge.search
ticket.create requires approval
}
on message {
search knowledge with message
reply using context
}
deploy {
web
api
slack
}
}Platform map