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.

Guardian

Guardian is a lightweight Node.js process that runs alongside your agent. It has no AI, uses zero tokens, and has one job: keep the agent online and let you manage it from chat.

What it watches

Every 30 seconds, Guardian:

Runs tmux capture-pane -p against the agent's session
Scans the captured output for error patterns (API Error:, Overloaded, Could not process, etc.)
Looks for incoming messages tagged ← discord, ← telegram, etc. to know the agent is receiving traffic
Checks for "working" indicators (Baked, Crunching, tool-use markers) to distinguish stuck from idle

If the session is dead, it restarts. If errors are piling up and no work is happening, it flags the agent as stuck and restarts after one additional failed check.

Restart policy

Max 3 restarts per 30 minutes. Past that, Guardian stops trying and pages you via your channel.
Every restart is logged to ~/workspace/memory/guardian.log and state written to /tmp/sento-guardian-<agent>.json.
On clean restart, Guardian sends a ✅ notification: "agent is back online."

Channel paging

Guardian detects which channel your agent uses (Discord, Telegram, or Slack), reads the bot token from the plugin's .env file, and uses the appropriate API to message you directly:

Discord: webhook if configured, otherwise Bot API to the monitor channel
Telegram: sendMessage to your monitor chat ID
Slack: chat.postMessage to your monitor channel

Interactive commands from chat

Guardian also polls the monitor channel for user messages, so you can manage the agent without SSH:

restart — Guardian stops and restarts the agent, resets restart count
logs — dumps the last 15 lines of tmux output back to the channel
status — prints health summary (alive / stuck / restart count / state)

Why no AI

Because anything with AI can fail. Guardian's whole point is to be the thing that's still alive when the agent isn't. Node.js, regex, tmux capture-pane. That's the whole stack.

Files to know

src/templates/guardian.js in the Sentō repo — the Guardian template (rendered per-agent)
~/workspace/memory/guardian.log — Guardian's log output
/tmp/sento-guardian-<agent>.json — persistent state (restart list, status flag)
~/.claude/channels/<channel>/.env — read by Guardian at startup to pick up the token

How it starts and persists

On install: sento init starts Guardian via nohup node guardian.js & and drops a @reboot cron line so it comes back on restart.
On reboot: the cron entry re-runs the start script, Guardian launches, agent launches.
No systemd, no launchd by default — nohup + tmux + cron is the simplest reliable pattern across distros and containers.