OpenClaw
Run a personal AI agent, persistent, private, and connected to your preferred channels.
Featherless Managed OpenClaw is a fully managed agent environment with sandboxed compute and inference bundled into a single monthly subscription. You get a secure, always-on OpenClaw instance with Featherless already wired as the model backend. No infrastructure to manage. No separate token bill.
OpenClaw handles the agent layer: persistent memory, skills, channels, and the gateway that connects everything.
Featherless handles the model layer: 27,000+ open-source models, no prompt logging, flat-rate pricing.
Deploy
Go to featherless.ai/account/agents and click Browse Marketplace. Find OpenClaw and click Launch. You'll land on a Launch OpenClaw screen with two fields:
Name (optional): a friendly label for this agent, e.g. "My Assistant"
Model: a searchable dropdown across all 27,000+ models on Featherless
The default model is MiniMaxAI/MiniMax-M2.5. Not sure what to pick? Here are a few good starting points:
Model | Good for |
|---|---|
| General-purpose default, balanced speed and quality |
| Reasoning-heavy tasks, long context |
| Complex tasks, large MoE model |
| Fast everyday assistant |
| Demanding tasks, massive MoE |
Hit Launch Agent. Your sandbox spins up in seconds and your gateway dashboard opens automatically at:
https://18789-sbx-[your-id].sandbox.featherless.ai
Dashboard
Your gateway dashboard is the control centre for everything. The status indicator in the top-left shows Health: OK when the gateway is running. If it shows anything else, refresh and wait 30 seconds.
Chat — talk directly to your agent in the browser. Backed by the same model and memory as any connected channel. The fastest way to test changes before rolling them out.
Agents — lists your configured agents. Click any agent, then open the Files tab to edit SOUL.md and MEMORY.md inline.
Config — your live openclaw.json. Switch to Raw mode (bottom-right toggle) to edit JSON directly. The save flow is always: edit → Save → Apply.
Channels — status snapshots for each connected channel. Shows whether a channel is configured, running, and connected.
Cron Jobs — lists all scheduled tasks. Toggle individual jobs on or off from here.
Logs — live and historical gateway logs. First place to check if a channel isn't responding or a config change didn't behave as expected.
Config vs dashboard UI for channels: The Channels tab provides a visual editor for channel settings. It's still early and some fields don't always save reliably. Use Config → Raw for channel configuration instead.
First conversation
In the dashboard, click Chat in the sidebar. Send a message to introduce yourself:
| "Hey, I'm [your name]. I work in [field]. I prefer concise answers and my timezone is CET."
Your agent replies within a few seconds and takes notes in MEMORY.md automatically.
Try these while you're here:
Web search
"What are the most downloaded open-source LLMs on HuggingFace this week?"
The agent has web_search available by default. It pulls live results, not training data.
URL summary
"Can you summarise this for me? [paste any URL]"
Connect Telegram
Telegram is the cleanest channel to start with. One token, one config block, done. Once set up, your agent is reachable from your phone without opening the dashboard.
1. Create a bot in BotFather
Open Telegram and search for @BotFather. Start a chat and send:
/newbot
Give your bot a display name (e.g. "My Assistant") and a username ending in bot (e.g. myassistant_bot). BotFather returns a token that looks like 7234891234:AABB.... Copy it.
2. Find your numeric Telegram user ID
Send any message to @userinfobot on Telegram. It replies instantly with your numeric ID — a plain number, not your username.
3. Configure Telegram
Option A: Config → Raw (recommended)
Go to Config → Raw in the dashboard. Add a "channels" block at the top level:
"channels": {
"telegram": {
"enabled": true,
"botToken": "YOUR_BOTFATHER_TOKEN",
"dmPolicy": "allowlist",
"allowFrom": ["YOUR_NUMERIC_TELEGRAM_USER_ID"],
"groups": {
"*": { "requireMention": true }
}
}
}Hit Save, then Apply.
Option B: Dashboard UI
Go to Channels → Telegram in the dashboard sidebar.
Paste your BotFather token into the Telegram Bot Token field
Under Telegram DM Policy, select allowlist
Under Allow From, click Add and enter your numeric user ID
Toggle Enabled on
Hit Save, then Reload
4. Test
Open Telegram and send your bot a message:
"Hey, it's me — same conversation, new channel."
You should get a response within a few seconds. If nothing comes back, check the Logs section in the dashboard.
DM policy options:
Policy | Behaviour |
|---|---|
| Unknown senders receive a verification code before messages are processed |
| Only senders in |
| Anyone can message the bot. Requires |
| Bot does not respond to any DMs |
Personalise your agent
SOUL.md
SOUL.md is your agent's system prompt. It loads into every session in full, on every channel. Use it to define who the agent is, what it does, and how it responds.
Go to Agents → [Your Agent] → Files → SOUL.md to edit it inline.
# Who you are
You're a personal assistant running on an open-source model via Featherless.
You're direct, concise, and useful. You don't pad responses with filler.
# What you do
- Answer questions using web search when needed
- Remember what I tell you across sessions (check MEMORY.md)
- Summarise links I send you
- Remind me of things I ask you to track
# How you respond
Keep answers short unless I ask for depth. If you're uncertain, say so.
Use plain language, not markdown formatting, when responding in Telegram.Keep SOUL.md under 2,000 words. Longer reference content belongs in MEMORY.md.
MEMORY.md
MEMORY.md is your agent's persistent notepad. After conversations, the agent writes notes here automatically: your name, preferences, things you've mentioned. Every new private session loads this file at the start.
Go to Agents → [Your Agent] → Files → MEMORY.md to read or edit it directly.
MEMORY.md is loaded in private (DM) sessions only, not group chats. Rules that need to apply everywhere belong in SOUL.md.
To test it: start a fresh conversation in Telegram without giving any context and ask:
"What do you know about me?"
Recurring reminders
Tell your agent directly in chat:
"Every morning at 8am, message me in Telegram with a motivational thought."
This creates a cron job. The agent messages you unprompted at that time each day. Manage it under Cron Jobs in the dashboard sidebar.
Troubleshooting
Bot doesn't respond in Telegram. Confirm allowFrom contains your numeric user ID, not your username. Send a message to @userinfobot to verify, update the config, and hit Apply.
Config changes not taking effect. Always hit Apply after saving. If the issue persists, use Reload to resync from disk.
Health shows something other than OK. Refresh the dashboard. If it persists, wait 30 seconds — the sandbox may be restarting.
spawn docker ENOENT error. The Featherless sandbox does not have Docker available. Set sandbox.mode to "off" if this appears in your config.
Channel configured but not responding. Confirm plugins.entries.[channel].enabled is true. A channel in channels but missing from plugins will not start.
Resources
OpenClaw GitHub: github.com/openclaw/openclaw
OpenClaw Docs: docs.openclaw.ai
Telegram channel docs: docs.openclaw.ai/channels/telegram
Featherless Discord: discord.gg/featherlessai