ContextQMD
Back to Openclaw

Personal assistant setup

docs/start/openclaw.md

2026.6.119.2 KB
Original Source

OpenClaw is a self-hosted gateway that connects Discord, Google Chat, iMessage, Matrix, Microsoft Teams, Signal, Slack, Telegram, WhatsApp, Zalo, and more to AI agents. This guide covers the "personal assistant" setup: a dedicated WhatsApp number that behaves like your always-on AI assistant.

⚠️ Safety first

You're putting an agent in a position to:

  • run commands on your machine (depending on your tool policy)
  • read/write files in your workspace
  • send messages back out via WhatsApp/Telegram/Discord/Mattermost and other bundled channels

Start conservative:

  • Always set channels.whatsapp.allowFrom (never run open-to-the-world on your personal Mac).
  • Use a dedicated WhatsApp number for the assistant.
  • Heartbeats now default to every 30 minutes. Disable until you trust the setup by setting agents.defaults.heartbeat.every: "0m".

Prerequisites

  • OpenClaw installed and onboarded - see Getting Started if you haven't done this yet
  • A second phone number (SIM/eSIM/prepaid) for the assistant

The two-phone setup (recommended)

You want this:

mermaid
flowchart TB
    A["<b>Your Phone (personal)
</b>
Your WhatsApp
+1-555-YOU"] -- message --> B["<b>Second Phone (assistant)
</b>
Assistant WA
+1-555-ASSIST"]
    B -- linked via QR --> C["<b>Your Mac (openclaw)
</b>
AI agent"]

If you link your personal WhatsApp to OpenClaw, every message to you becomes "agent input". That's rarely what you want.

5-minute quick start

  1. Pair WhatsApp Web (shows QR; scan with the assistant phone):
bash
openclaw channels login
  1. Start the Gateway (leave it running):
bash
openclaw gateway --port 18789
  1. Put a minimal config in ~/.openclaw/openclaw.json:
json5
{
  gateway: { mode: "local" },
  channels: { whatsapp: { allowFrom: ["+15555550123"] } },
}

Now message the assistant number from your allowlisted phone.

When onboarding finishes, OpenClaw auto-opens the dashboard and prints a clean (non-tokenized) link. If the dashboard prompts for auth, paste the configured shared secret into Control UI settings. Onboarding uses a token by default (gateway.auth.token), but password auth works too if you switched gateway.auth.mode to password. To reopen later: openclaw dashboard.

Give the agent a workspace (AGENTS)

OpenClaw reads operating instructions and "memory" from its workspace directory.

By default, OpenClaw uses ~/.openclaw/workspace as the agent workspace, and will create it (plus starter AGENTS.md, SOUL.md, TOOLS.md, IDENTITY.md, USER.md, HEARTBEAT.md) automatically on setup/first agent run. BOOTSTRAP.md is only created when the workspace is brand new (it should not come back after you delete it). MEMORY.md is optional (not auto-created); when present, it is loaded for normal sessions. Subagent sessions only inject AGENTS.md and TOOLS.md.

<Tip> Treat this folder like OpenClaw's memory and make it a git repo (ideally private) so your `AGENTS.md` and memory files are backed up. If git is installed, brand-new workspaces are auto-initialized. </Tip>
bash
openclaw setup

Full workspace layout + backup guide: Agent workspace Memory workflow: Memory

Optional: choose a different workspace with agents.defaults.workspace (supports ~).

json5
{
  agents: {
    defaults: {
      workspace: "~/.openclaw/workspace",
    },
  },
}

If you already ship your own workspace files from a repo, you can disable bootstrap file creation entirely:

json5
{
  agents: {
    defaults: {
      skipBootstrap: true,
    },
  },
}

The config that turns it into "an assistant"

OpenClaw defaults to a good assistant setup, but you'll usually want to tune:

  • persona/instructions in SOUL.md
  • thinking defaults (if desired)
  • heartbeats (once you trust it)

Example:

json5
{
  logging: { level: "info" },
  agents: {
    defaults: {
      model: { primary: "anthropic/claude-opus-4-6" },
      workspace: "~/.openclaw/workspace",
      thinkingDefault: "high",
      timeoutSeconds: 1800,
      // Start with 0; enable later.
      heartbeat: { every: "0m" },
    },
    list: [
      {
        id: "main",
        default: true,
        groupChat: {
          mentionPatterns: ["@openclaw", "openclaw"],
        },
      },
    ],
  },
  channels: {
    whatsapp: {
      allowFrom: ["+15555550123"],
      groups: {
        "*": { requireMention: true },
      },
    },
  },
  session: {
    scope: "per-sender",
    resetTriggers: ["/new", "/reset"],
    reset: {
      mode: "daily",
      atHour: 4,
      idleMinutes: 10080,
    },
  },
}

Sessions and memory

  • Session files: ~/.openclaw/agents/<agentId>/sessions/{{SessionId}}.jsonl
  • Session metadata (token usage, last route, etc): ~/.openclaw/agents/<agentId>/sessions/sessions.json (legacy: ~/.openclaw/sessions/sessions.json)
  • /new or /reset starts a fresh session for that chat (configurable via resetTriggers). If sent alone, OpenClaw acknowledges the reset without invoking the model.
  • /compact [instructions] compacts the session context and reports the remaining context budget.

Heartbeats (proactive mode)

By default, OpenClaw runs a heartbeat every 30 minutes with the prompt: Read HEARTBEAT.md if it exists (workspace context). Follow it strictly. Do not infer or repeat old tasks from prior chats. If nothing needs attention, reply HEARTBEAT_OK. Set agents.defaults.heartbeat.every: "0m" to disable.

  • If HEARTBEAT.md exists but is effectively empty (only blank lines, Markdown/HTML comments, Markdown headings like # Heading, fence markers, or empty checklist stubs), OpenClaw skips the heartbeat run to save API calls.
  • If the file is missing, the heartbeat still runs and the model decides what to do.
  • If the agent replies with HEARTBEAT_OK (optionally with short padding; see agents.defaults.heartbeat.ackMaxChars), OpenClaw suppresses outbound delivery for that heartbeat.
  • By default, heartbeat delivery to DM-style user:<id> targets is allowed. Set agents.defaults.heartbeat.directPolicy: "block" to suppress direct-target delivery while keeping heartbeat runs active.
  • Heartbeats run full agent turns - shorter intervals burn more tokens.
json5
{
  agents: {
    defaults: {
      heartbeat: { every: "30m" },
    },
  },
}

Media in and out

Inbound attachments (images/audio/docs) can be surfaced to your command via templates:

  • {{MediaPath}} (local temp file path)
  • {{MediaUrl}} (pseudo-URL)
  • {{Transcript}} (if audio transcription is enabled)

Outbound attachments from the agent use structured media fields on the message tool or reply payload, such as media, mediaUrl, mediaUrls, path, or filePath. Example message-tool arguments:

json
{
  "message": "Here's the screenshot.",
  "mediaUrl": "https://example.com/screenshot.png"
}

OpenClaw sends structured media alongside the text. Legacy final assistant replies may still be normalized for compatibility, but tool output, browser output, streaming blocks, and message actions do not parse text as attachment commands.

Local-path behavior follows the same file-read trust model as the agent:

  • If tools.fs.workspaceOnly is true, outbound local media paths stay restricted to the OpenClaw temp root, the media cache, agent workspace paths, and sandbox-generated files.
  • If tools.fs.workspaceOnly is false, outbound local media can use host-local files the agent is already allowed to read.
  • Local paths can be absolute, workspace-relative, or home-relative with ~/.
  • Host-local sends still only allow media and safe document types (images, audio, video, PDF, Office documents, and validated text documents such as Markdown/MD, TXT, JSON, YAML, and YML). This is an extension of the existing host-read trust boundary, not a secret scanner: if the agent can read a host-local secret.txt or config.json, it can attach that file when the extension and content validation match.

That means generated images/files outside the workspace can now send when your fs policy already allows those reads, while arbitrary host-local text extensions remain blocked. Keep sensitive files outside the agent-readable filesystem, or keep tools.fs.workspaceOnly=true for stricter local-path sends.

Operations checklist

bash
openclaw status          # local status (creds, sessions, queued events)
openclaw status --all    # full diagnosis (read-only, pasteable)
openclaw status --deep   # asks the gateway for a live health probe with channel probes when supported
openclaw health --json   # gateway health snapshot (WS; default can return a fresh cached snapshot)

Logs live under /tmp/openclaw/ (default: openclaw-YYYY-MM-DD.log).

Next steps