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.

CLI reference

After running npx sentoagent init once, the sento binary is globally installed and in your PATH (you may need to source ~/.bashrc or start a new shell). Everything after setup uses sento <command>.

Setup

npx sentoagent init

One-time. Installs Claude Code, the chosen channel plugin, ClawMem, Guardian, the cron entries, and launches the agent. Asks a few questions interactively.

Process control

sento status           # Is the agent alive? Linked? How many restarts?
sento start            # Start the agent (tmux session + Guardian)
sento stop             # Stop cleanly (SIGTERM, graceful)
sento restart          # stop + start

sento status prints agent health, channel connection state, Guardian restart count for the current window, and the last heartbeat timestamp.

Configuration

sento config           # Change model, OAuth token, channel tokens, keys
sento channels         # Add / remove / list channels on this agent
sento channels add discord
sento channels remove telegram

sento config walks you through the same prompts as init, but only re-asks the keys you want to change.

Observability

sento logs             # Last 100 lines of agent output (tmux capture)
sento logs -n 500      # Last 500 lines
sento logs --watchdog  # Guardian's own log, separate from the agent

Skills

sento skills           # List installed skills
sento skills install <path>
sento skills export <name>
sento skills import <tarball>

Skills are Markdown-based extensions that teach your agent new capabilities. Share them between your agents or publish them as gists.

Maintenance

sento update           # Update CLI + Claude Code + plugins + re-apply patches
sento doctor           # Run 12 health checks
sento doctor --fix     # Run checks and auto-fix what can be fixed

sento doctor checks: Node version, Claude Code installed, OAuth token present, tmux session alive, Guardian process alive, cron entries installed, channel plugin patched, access.json valid JSON, disk space, memory, and more.

Agent-to-agent

sento pair             # Start pairing with another agent (generates code)
sento agents           # List this agent's info and paired agents

Pairing is interactive: one agent runs sento pair, shares the SENTO-XXXX code, the other agent owner types it in, both agents exchange an HMAC secret, and from then on they can message each other over HTTP. See agent-to-agent.

Everything the CLI does, the agent can do

When you install Sentō, the agent's CLAUDE.md includes a "self-administration" section that authorizes it to run any sento command on itself. So instead of SSHing in, you can message the agent:

"restart yourself"

"add telegram as a channel"

"run doctor and tell me what's broken"

It'll do it. This is documented under self-administration.