setupagents
Deploy an agent
NEW Managed OpenClaw deployments are live

Infrastructure that keeps
your agents online.

Deploy Hermes and OpenClaw agents to Telegram in minutes. Managed servers, persistent memory, observability, and updates—all in one place.

Deploy your first agent Explore the platform
No server setup Live in under two minutes
Agentshermes-telegram
FRA-1
TELEGRAM AGENT

Hermes assistant

Live
Uptime 24H100%All systems healthy
Messages 24H1,284+18.2% from yesterday
Response time MEDIAN1.2s-240ms this week
Memory USED38%2.1 GB of 6 GB
MESSAGE ACTIVITY

Agent traffic

12 AM6 AM12 PM6 PMNOW
QUICK DEPLOY

Hermes → Telegram

A capable personal agent with memory, tools, and a fast Telegram loop.

agentctl deploy hermes --channel telegram
→Select a runtime and connect Telegram.
→Configuration is ready for deployment.

One platform for every Telegram agent

99.99%PLATFORM UPTIME
<2 minAVERAGE DEPLOY
24 / 7HEALTH MONITORING
1-clickSAFE ROLLBACKS
BUILT FOR TELEGRAM

The secure rails for
every kind of agent

From one private assistant to a public community bot, each deployment gets production infrastructure by default.

Community assistant

Answer questions, search knowledge, moderate chats, and keep your Telegram community moving.

Explore

Personal operator

Give yourself a persistent assistant for research, reminders, writing, and daily workflows.

Explore

Workflow automation

Connect APIs, trigger jobs, watch inboxes, and turn Telegram messages into real actions.

Explore

Research agent

Monitor sources, summarize changes, and deliver concise intelligence into a private chat.

Explore

Support agent

Resolve repeat questions instantly and route higher-stakes conversations to a human.

Explore

Custom agent

Bring your own system prompt, skills, model keys, and repository without changing your runtime.

Explore
THE COMPLETE AGENT PLATFORM

Everything between your
prompt and Telegram

Use the defaults and deploy immediately, or control every part of the runtime as your agent grows.

RUNTIME

Hermes and OpenClaw,
ready out of the box

Current releases, tuned defaults, persistent workspaces, and the tools each framework expects.

Compare runtimes
HermesRunning · v3
OpenClawReady · stable
TelegramWebhook healthy
CONTROL

Full terminal access

Inspect logs, update configuration, install skills, and debug without leaving your browser.

agent-shell
$ agentctl status
● runtime   healthy
● telegram  connected
● memory    mounted

$
MEMORY

State that survives

Conversations, files, schedules, and long-term memory stay attached through restarts and upgrades.

Agent workspaceEncrypted & persistent
OPERATIONS

Managed infrastructure that stays invisible

Automatic restarts, encrypted backups, version pinning, private networking, and live health checks are included with every deployment.

View plans
Health checksEvery 30s
Encrypted secretsAt rest + transit
Daily backup7 day retention
Global edge8 regions
RELIABLE BY DEFAULT

Focus on the agent.
We'll run the rest.

A small, complete operations layer for software that needs to answer every message.

One-click deployment

Choose Hermes or OpenClaw, paste a Telegram bot token, and go live on managed compute.

Isolated runtimes

Every agent gets dedicated containers, encrypted storage, and strict network boundaries.

Persistent memory

Agent history, workspace files, and long-term memory survive restarts and upgrades.

Logs and health

See uptime, response latency, token usage, process health, and recent conversations in real time.

Secret management

Store model keys and Telegram credentials encrypted, then rotate them without a redeploy.

Safe upgrades

Pin a version, stage an update, or roll back to the last working release in one click.

PRICING

Start small.
Scale when it works.

Every plan includes Telegram delivery, managed HTTPS, persistent storage, logs, and automatic restarts.

Hobby

A private assistant or low-traffic bot.

$19/ monthStart deploying
  • 1 agent
  • 2 vCPU · 2 GB RAM
  • 5 GB persistent storage
  • 3-day log history
MOST POPULAR

Pro

An always-on agent used every day.

$39/ monthStart deploying
  • 3 agents
  • 4 vCPU · 8 GB RAM
  • 40 GB persistent storage
  • 30-day logs + backups

Scale

Public bots, communities, and teams.

$99/ monthStart deploying
  • 10 agents
  • 8 vCPU · 16 GB RAM
  • 160 GB persistent storage
  • Priority support + SSO
FAQ

From bot token
to online agent

Everything you need to know before your first deployment.

Read the documentation

A Telegram bot token from BotFather and an API key for the model provider you want to use. Choose Hermes or OpenClaw, add those secrets, and the platform handles the server, runtime, HTTPS, health checks, and updates.

Yes. Import your configuration, prompt, skills, workspace files, and environment variables. You can also connect a Git repository and keep your current project structure.

Yes. Agents run on always-on managed infrastructure with automatic restarts, health checks, and persistent storage. Sleep mode is available on lower plans when you want to reduce cost.

You control the allowlist. Keep an agent private to your Telegram account, open it to selected users or groups, or make it publicly available. Rate limits and moderation rules can be configured per deployment.

Yes. Bring keys for the providers supported by your agent runtime, or use an OpenAI-compatible endpoint. Model credentials remain encrypted and can be rotated at any time.

YOUR AGENT IS READY

Build something
worth messaging.

Deploy Hermes or OpenClaw to Telegram in under two minutes.