Clawd Console
Your command deck

Clawdbot User Guide (Install + Setup)

This is the single, operator-friendly Clawdbot user guide that combines the three intents people actually have: a Clawdbot install guide, a practical Clawdbot setup walkthrough, and a clear path into troubleshooting when something breaks.

Quick start checklist (10 minutes)

  1. Pick your install mode: VPS/systemd (recommended for operators) or Docker (good for labs).
  2. Decide One-Brain: one service user, one HOME, one DATA_DIR, one set of ports.
  3. Start the gateway + console and confirm the UI loads.
  4. Connect your first channel (Signal/Telegram/etc.) and send a test message.
  5. If you need browser automation: start headless Chrome with a stable CDP port and persistent profile.
If you read one thing about reliability
Most “random” failures come from split state (root vs non-root, duplicate services, port collisions). Read the One-Brain series:

Where to go next (fast links)

Image placeholder
(Insert: screenshot of the Console home with gateway status + the PM board.)

Install guide (VPS/systemd) — recommended path

For long-lived bots, systemd is the most reliable baseline: it makes restarts predictable, logs accessible, and process ownership obvious.

  1. Create a dedicated service user (example: master).
  2. Choose your workspace (example: /home/master/clawd).
  3. Make HOME + WorkingDirectory explicit in your service unit(s).
  4. Pick a durable DATA_DIR for transcript/PM/worklogs (and back it up).
  5. Start services and verify ports/listeners.
# Example shape (conceptual)
[Service]
User=master
Group=master
WorkingDirectory=/home/master/clawd
Environment=HOME=/home/master
Environment=DATA_DIR=/home/master/clawd/apps/console-data
Restart=on-failure

If you’re already installed and seeing instability, jump straight to the reliability runbook: One Brain, One Console.

Install guide (Docker)

Docker is great for isolated environments and reproducibility. The #1 rule is still One-Brain: mount durable volumes so state survives container restarts.

  • Mount persistent volumes for:
    • the bot config/home (auth profiles)
    • DATA_DIR (console data, transcript, PM)
    • optional: Chrome profile directory if running headless in-container
  • Expose only required ports and avoid overlapping ports with other stacks.

If Docker “works but isn’t stable,” the root cause is usually missing volumes (state resets) or port collisions.

Setup (auth, secrets, first run)

After install, setup is about making sure your bot has stable identity and stable permissions.

  • Auth profiles: configure your provider profiles in the correct HOME for the service user.
  • Secrets: keep tokens in env files / locked-down config, not in command history.
  • First run: send a test message, verify responses, and confirm transcript/PM writes land in your intended DATA_DIR.
  • Browser automation (optional): run Chrome with a stable CDP port and persistent --user-data-dir to keep logins.
Browser help:

Troubleshooting (links to FAQ cluster)

If something broke, use the hub page and jump to the most specific symptom page:

If issues feel nondeterministic, read the One-Brain series first — it resolves the majority of “ghost problems.”

Get Early Access!

Launching any day…

Hosted Clawdbot + ClawdConsole workspaces are coming very, very soon.

Join the waitlist and we’ll spin up your workspace the moment access opens.

  • Same operator cockpit you’re seeing here
  • Private hosted workspace
  • We’ll email you your link when it’s ready