Skip to main content
Use artil init to have incoming messages notify an open Claude Code session or start a Hermes run. First install the CLI, sign in as your agent, and check that you can read its email. For inbox checks during a conversation, use the Claude Code plugin or Pi extension.

​
1. Choose where the agent will run

Choose a client below. Local setup adds message delivery, the artil skill, secret loading, and the agent’s MCP connections. Choosing a new client replaces the existing delivery setup, including any remote Hermes webhook for this agent.
Local setup links all of the agent’s stored secrets. Members who can access a workspace agent can reveal those secrets. The Claude Code link uses your user-level configuration and applies across projects; Hermes uses its configured installation. Review the agent’s secrets before connecting a client.
Install Claude Code and use Anthropic authentication. Channels are not available through Bedrock, Google Cloud’s Agent Platform, or Microsoft Foundry. On Team and Enterprise plans, an owner must enable channels first. Console organizations that deploy managed settings must enable channelsEnabled. The development flag below cannot override a disabled channel setting. See Claude’s channel requirements.Connect this computer:
Setup adds the artil-messages channel and allows the CLI commands that read and send messages. Complete any steps under STILL YOURS TO DO, such as fixing a skill conflict or finishing an MCP connection.Start a new Claude Code session using the command setup prints:
Check that Claude starts with the channel enabled. Resolve any startup warning about blocked channels. The new session also loads the secrets hook. Keep it open for the email check below; closing it stops notifications.

​
2. Check an incoming message

From your own email account, send a new message to the agent’s address. For example: “Please reply to confirm you can read this email.” The client receives a notification containing the message ID. Local clients use the CLI to read and reply; the remote Hermes webhook prompts it to use its configured message tools. Watch the client read the message, then check your inbox for its reply.
Keep Claude’s session open, or check the background-job log for local Hermes. For a remote Hermes server, check that webhook setup is complete and the URL is reachable. Inspect any client permission prompts before expecting a reply.Run artil auth status to check the agent and its channels. Sender restrictions, spam filtering, and notification limits can stop a message from reaching the client. See message troubleshooting.
Claude may show the same notification after reconnecting or in another session. Keep one answering session open per agent to reduce duplicate replies.

Next: sign up for a service

Use this inbox to receive a verification email, finish signup, and store the service’s credentials.

​
Change or disconnect a client

Run artil init to choose a different client. A new choice replaces the existing delivery setup, including the agent’s remote Hermes webhook if you switch to a local client. Keeping the same choice repairs outdated setup and updates the installed skill, secret links, and MCP entries.If a separately installed skill has the same name, setup keeps it and reports the conflict under STILL YOURS TO DO.
This command removes delivery setup on this computer and the selected agent’s remote Hermes webhook. It stops webhook notifications to that server even if another computer configured the connection.
It also removes the skills, secret links, and MCP entries that setup added for the disconnected local clients. Your agent, messages, and secrets remain stored in Artil. The webhook route in Hermes’ own configuration stays in place; you can remove that route yourself.Remove separately linked secrets, such as a project’s .env, with artil secrets unlink <place>. Stop programs that already loaded those values. For CLI sign-out, see manage a login.