Does Sentō need a pay-per-token API key?

No. Sentō authenticates using the same Claude Code OAuth token you already sign in with on your machine, and that token is tied to your Claude Pro or Max subscription, not to a pay-as-you-go Anthropic Console key. Every message your agent sends counts against your subscription, not a separate metered invoice. If you prefer to run on a pay-per-token API key instead, Sentō supports that too.

How much does it cost to run a Sentō agent 24/7?

Zero extra on top of the Claude subscription you're already paying for. A Pro plan is enough for a daily-driver agent; Max unlocks heavier 24/7 workloads. Sentō also routes everything through Claude Code, which means you get Claude Code's own context and caching optimizations, so the same token goes further than it would through a raw API call. You never see a separate token bill.

How long does setup take?

About five minutes. Run `npx sentoagent init`, answer a few questions about model, channel, and bot token, and the agent launches and comes online on Discord, Telegram, Slack, or iMessage.

What makes Sentō self-healing?

Every agent runs under a Guardian + Watchdog pair. Guardian restarts the agent process if it crashes; Watchdog watches Guardian. If the agent goes down, it's back up before you notice, and you can fix anything right from Discord without SSH'ing in.

Can two Sentō agents talk to each other?

Yes. Sentō is the first framework with built-in, secure agent-to-agent messaging. Every message is HMAC-SHA256 signed, pairing requires mutual consent on both sides, and rate limiting is on by default.

Which messaging platforms are supported?

Discord, Telegram, Slack, and iMessage out of the box. You can run on a VPS, inside Docker, or directly on macOS.

Is Sentō open source?

Yes, MIT licensed. The source lives at github.com/sentoagent/sento and the package at npmjs.com/package/sentoagent.

Quickstart

Sentō takes one command. Prerequisites: Node 20+ and a Claude subscription (Pro or Max).

npx sentoagent init

The installer will ask you:

Agent name — pick anything. It's the name you'll call it by.
Channel — Discord, Telegram, Slack, or iMessage.
Channel token — Bot token for Discord / Telegram / Slack. iMessage skips this (it reads the local Messages database on macOS).
Model / personality / timezone — a few quick prompts to configure the agent's identity.
Authentication — a URL appears in your terminal. Open it on any device (phone or laptop), log in with your Claude account, and paste the authorization code back. No token to prepare beforehand.

About five minutes later the agent is running in a tmux session, the Guardian is watching it, and it's online in your chosen channel.

First message

Send your agent any message in a channel it can see. A few things happen:

It reacts with 👀 to confirm it received the message.
It starts an onboarding conversation — it asks about your role, how you want it to talk, what it should help with.
After onboarding, it writes what it learned to its own CLAUDE.md identity file. This is how it remembers you across sessions.

Once onboarded, try:

hey, what's the weather in Tokyo? — browses the web via Playwright.
write a python script that reads a CSV and plots it — file + code operations.
remember that I prefer concise answers — persistent memory via ClawMem.
every day at 9am, DM me the top 3 HN headlines — scheduled tasks.

Common commands after install

sento status          # check agent health
sento logs -n 50      # view last 50 lines of output
sento restart         # restart the agent in tmux
sento doctor --fix    # diagnose and auto-fix issues

Full CLI surface: CLI reference.

Where things live

Agent workspace: ~/workspace/ on the host. Contains CLAUDE.md, start script, memory folder.
Channel config: ~/.claude/channels/<channel>/.env — bot tokens, access control.
Guardian state: /tmp/sento-guardian-<agent>.json — restart count, health state.

Next steps

Full VPS install walkthrough if you want the agent to live somewhere other than your laptop.
Read about the Guardian — the process that keeps your agent from ever going offline silently.
Try agent-to-agent pairing once you have two agents running.