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.

Telegram channel

Telegram is the fastest channel to set up — about 90 seconds. Bots are private by default, so no one can find yours unless you share its username.

Create the bot

Open Telegram, search for @BotFather, start a chat.
/newbot
Display name (e.g. "Mira Agent")
Username ending in _bot (must be globally unique, e.g. mira_agent_bot)
BotFather sends a token like 123456789:ABCdef…. Save it.

Open a chat with your bot

BotFather's message includes t.me/<username>. Click it, open a DM with your bot, press Start. The bot stays silent until Sentō is running — this just creates the chat so the agent can DM you later.

Run sento init

npx sentoagent init

Pick Telegram, paste the token. No server ID required — Telegram bots work per-chat via chat_id.

Access control

~/.claude/channels/telegram/.env stores the token. Access rules live in ~/.claude/channels/telegram/access.json:

{
  "dmPolicy": "allowlist",
  "allowFrom": ["123456789"],
  "groups": {
    "-1001234567890": { "requireMention": true, "allowFrom": [] }
  }
}

allowFrom for DMs holds Telegram user IDs (numeric).
groups keys are chat IDs (group chats are negative numbers starting with -100).
requireMention: true in a group means the bot only responds when you @mention its username.

Finding a chat or user ID

Easiest: send a message to the bot, then query:

curl "https://api.telegram.org/bot<TOKEN>/getUpdates"

Every message payload includes message.chat.id and message.from.id.

Why Telegram is quick

No intents, scopes, or developer portal.
Bots work out of the box with Markdown formatting.
long-polling (via getUpdates) works without a webhook URL, so Telegram runs fine on residential IPs with no port forwarding.