v2026.4.5 · Updated Apr 6, 2026350K+ GitHub stars16 sections

The OpenClaw Cheatsheet

Every command, config, file, and trick — on one page. Bookmark this.

More assistant cheatsheets📎 Paperclip Hermes Agent
✨ Workspace Wizard NEW⚙️ Config Builder NEW🧩 Skill Explorer NEW💰 Cost Calculator NEW📄 Download cheatsheet (print / PDF)Share on 𝕏⭐ Star OpenClaw

Want to support this page and keep it free with the latest tips & tricks?
☕ Buy me a coffee · other ways to support

OpenClaw in 60 seconds

✅ Latest 2 Release Migration Checklist

Run these once after upgrading to quickly align config and verify the biggest changes from v2026.4.5 and v2026.4.2.

openclaw doctor --fix      # Migrate legacy aliases + x_search/Firecrawl path moves
openclaw config validate   # Confirm canonical config is valid
openclaw plugins install --force <plugin-or-hook>  # v2026.4.5 safe replace flow
openclaw flows list       # v2026.4.2 Task Flow substrate sanity check
openclaw memory rem-harness # Optional: verify dreaming/REM pipeline (experimental)

v2026.4.5 · v2026.4.2 · all release notes

📡 Status Dashboard — Live Gateway Monitorv2026.4.5

Monitor your OpenClaw gateway in real time: channel health, memory index, skills, model costs, system resources. Connect to your live gateway or try Demo Mode — no setup required. Bookmark it.

📡 Open Status Dashboard

📚 Popular OpenClaw Guides

Looking for specific setup help? Start with these focused pages:

🚀 Zero to Running in 5 Minutes

1
curl -fsSL https://openclaw.ai/install.sh | bash
2
openclaw onboard --install-daemon
3
Generate workspace files with the Workspace Wizard NEW
4
openclaw dashboard → browser → talk to your lobster
5
Add Telegram or WhatsApp in the wizard for mobile
🆕

What's NewRecent releases

Highlights from the latest OpenClaw releases on GitHub — media tools, task flows, config migrations, and platform changes. Follow the links for curated feature pages and full notes.

v2026.4.5 Latest

FeatureDetails
Breaking: config aliasesLegacy keys removed — openclaw doctor --fix migrates; cliBackends dropped.
video_generate / music_generateBuilt-in media tools; xAI, Runway, Wan, Lyria, MiniMax, Comfy-backed workflows.
Comfy pluginLocal / Comfy Cloud workflows for image, video, and music generation.
Providers & MantleQwen, Fireworks, StepFun bundles; Bedrock Mantle + inference profiles + IAM bearer.
Control UI12+ locales; ClawHub search/install in Skills panel; plugins install --force.
/dreamingExperimental memory promotion, Dreams UI, dreams.md, REM tooling.
Claude CLI MCPOpenClaw tools via loopback MCP; stream-json partial streaming; env hardening.

Release notes →

v2026.4.2 Task flows

FeatureDetails
Breaking: xAI / Firecrawlx_search & Firecrawl web_fetch move under plugins.entries.*doctor --fix.
Task Flow substrateDurable state, managed children, api.runtime.taskFlow, openclaw flows recovery.
Android AssistantApp Actions — launch OpenClaw from Google Assistant into the composer.
before_agent_replyPlugin hook to short-circuit with a synthetic reply before the model.
Feishu commentsDrive comment events, thread context, in-thread replies.
CompactionConsistent agents.defaults.compaction.model; optional notifyUser.
Matrix m.mentionsReliable @mentions in Element-class clients.

Release notes →

Earlier releases

v2026.4.1/tasks, SearXNG, Bedrock Guardrails, cron --tools, agents.defaults.params. v2026.3.31flows, unified ledger, QQ Bot, Matrix/MCP. v2026.3.28config schema, Qwen migration, plugin approvals.

v2026.3.22 skills CLI → · All releases

Breaking changes (recent)

v2026.4.5

  • Legacy config aliases removed — migration via openclaw doctor --fix
  • Bundled CLI text backends and agents.defaults.cliBackends removed

v2026.4.2

  • x_searchplugins.entries.xai.config.xSearch.*; Firecrawl → plugins.entries.firecrawl.config.webFetch.*details

v2026.3.31

  • nodes.run removed — use exec host=node; install fail-closed; trusted-proxy / node-command rules

v2026.3.22

  • Chrome extension relay, legacy env prefixes — openclaw doctor --fix

Update Now CLI

sudo npm i -g openclaw@latest
openclaw update status
openclaw update --dry-run  # Preview first
openclaw update wizard
openclaw gateway restart

All releases

Quick Install

If you are searching for how to install OpenClaw quickly, this is the fastest path from zero to a working personal AI assistant. The commands below cover both one-line install and source install, then move you directly to onboarding, health checks, and first-run verification.

One-Liner CLI

# macOS / Linux
curl -fsSL https://openclaw.ai/install.sh | bash

# Windows (PowerShell)
iwr -useb https://openclaw.ai/install.ps1 | iex

# npm global
npm i -g openclaw

# Onboard wizard
openclaw onboard --install-daemon

From Source Advanced

git clone https://github.com/openclaw/openclaw.git
cd openclaw
pnpm install && pnpm ui:build && pnpm build
pnpm openclaw onboard --install-daemon

# Dev loop (auto-reload)
pnpm gateway:watch

Requirements

  • Node.js 22+ (node --version)
  • macOS 14+, Linux, or Windows (WSL2)
  • AI API key — Anthropic, OpenAI, OpenRouter, Mistral, or Google
  • Low RAM? export NODE_OPTIONS="--max-old-space-size=4096"

🍎 Full Mac Setup Guide NEW

Step-by-step walkthrough for macOS — Homebrew, Node.js, install, wizard, daemon, troubleshooting. Apple Silicon & Intel.

📖 Open Mac Guide

Post-Install

openclaw --version          # Version (includes git hash)
openclaw doctor              # Health check
openclaw doctor --deep --yes # Deep + auto-fix
openclaw test model          # Test AI
openclaw dashboard            # Web UI
openclaw message send --target +1555... --message "Hi"  # Test channel
📁

Workspace Files

OpenClaw behavior is mostly defined by workspace files, not by hidden prompts. Getting SOUL.md, AGENTS.md, USER.md, and TOOLS.md right gives you better results than model switching alone. Use this as your practical reference for what each file does and when it is loaded.

File Reference Files

All in ~/.openclaw/workspace/ — agent reads at session start.

FilePurposeLoaded
AGENTS.mdOperating instructions, priorities, workflow rulesEvery session
SOUL.mdPersonality, tone, values, behavioral constraintsEvery session
USER.mdAbout you — name, preferences, styleEvery session
IDENTITY.mdAgent name, role, goals, voiceEvery session
TOOLS.mdLocal tool notes, calendar IDs, conventionsEvery session
HEARTBEAT.mdChecklist for periodic heartbeat runsHeartbeat only
BOOT.mdStartup ritual on gateway restartBoot hook
BOOTSTRAP.mdFirst-run interview — auto-deleted afterFirst run
MEMORY.mdLong-term curated memory (optional)Main DM only
dreams.mdDreaming trail / diary when experimental dreaming is on (v2026.4.5+)Optional; inspectable
memory/YYYY-MM-DD.mdDaily logs — today + yesterday loadedSession start

What Goes Where Tip

  • AGENTS.md → stable rules, not temp tasks
  • SOUL.md → personality, not tickets
  • USER.md → personalization (tone, formatting)
  • TOOLS.md → guidance only, doesn't control access
  • MEMORY.md → curated facts, not transcripts

✨ Generate All Files Instantly NEW

Don't know what to write? The Workspace Wizard asks 8 questions and generates all your workspace files — SOUL.md, AGENTS.md, USER.md, IDENTITY.md, TOOLS.md, HEARTBEAT.md — ready to download.

✨ Open Workspace Wizard

⏰ Your First 24 Hours NEW

Just installed? This hour-by-hour guide walks you through SOUL.md, USER.md, first channel, cron jobs, heartbeat, security & model failover.

⏰ Open First 24h Guide

📋 Ready-to-Use Playbooks NEW

Complete workflows you can copy-paste: Developer Workflow (PR reviews, CI monitoring, dependency audits) and Personal Assistant (morning briefings, email triage, meal planning). Full configs, cron jobs & SOUL.md included.

📋 Browse Playbooks

Git Backup CLI

cd ~/.openclaw/workspace
git init && git add AGENTS.md SOUL.md TOOLS.md \
  IDENTITY.md USER.md HEARTBEAT.md memory/
git commit -m "Add workspace"

Never commit: API keys, OAuth tokens, config.

⚙️

Configuration openclaw.json

Most setup issues come from configuration mismatches: model choice, channel auth, heartbeat behavior, and security defaults. This section gives you copy-paste-safe openclaw.json patterns for stable day-one setup, plus fallback and reliability settings for long-running assistants.

Starter Config Config

// ~/.openclaw/openclaw.json — JSON5
{
  identity: { name: "Clawd", emoji: "🦞", theme: "helpful lobster" },
  agent: {
    workspace: "~/.openclaw/workspace",
    model: { primary: "anthropic/claude-sonnet-4-6" },
    heartbeat: { every: "30m", target: "last" },
  },
  channels: { telegram: { enabled: true, botToken: "123456:ABC..." } }
}

Config CLI CLI

openclaw config file           # Print active config path
openclaw config get <key>     # Values redacted in output
openclaw config set <key> <val>
openclaw config unset <key>
openclaw config validate       # Check JSON/schema before restart
openclaw config schema         # Print generated schema (v2026.3.28+)
openclaw gateway restart      # Required after changes

Model Fallbacks Config

agent: { model: {
  primary: "anthropic/claude-sonnet-4-6",
  fallbacks: ["anthropic/claude-opus-4-6", "mistral/mistral-large"],
  params: { context1m: true }  // 1M context beta
}}

Auto-rotates auth on rate limits. Per-agent params overrides supported.

Gateway Auth v2026.3.7+

When both gateway.auth.token and gateway.auth.password are set (including SecretRefs), you must set gateway.auth.mode to token or password. Required before upgrade.

gateway: {
  auth: {
    mode: "token",   // or "password"
    token: "..."
  }
}

⚙️ Build Your Config Visually NEW

Don't want to edit JSON manually? The Config Builder lets you pick models, channels, heartbeat, security — and generates a valid openclaw.json with live preview.

⚙️ Open Config Builder
⌨️

Core CLI Commands

The OpenClaw CLI is your primary interface for managing the gateway, channels, agents, and plugins. Every command below works in both interactive and scripted contexts, so you can use them in cron jobs, shell aliases, or CI pipelines. Memorize doctor, logs --follow, and gateway restart — they solve most day-to-day issues.

Gateway CLI

openclaw gateway status|start|stop|restart
openclaw dashboard         # Web UI
openclaw tui               # Terminal UI

Channels CLI

openclaw channels list
openclaw channels status --probe
openclaw channels login          # WhatsApp QR
openclaw channels add --channel telegram --token $TOK
openclaw channels logs --channel telegram
openclaw channels remove --channel discord --delete

Agents CLI

openclaw agents list --bindings
openclaw agents add|delete <name>
openclaw agents bindings|bind|unbind
openclaw agents set-identity --from-identity

Logs CLI

openclaw logs --follow
openclaw logs --json --limit 200
openclaw status --all --deep   # Full diagnosis
openclaw status --usage        # Provider usage/quota
openclaw doctor --deep --yes
openclaw health --json --verbose

Plugins & Skills CLI

openclaw plugins list|info|enable|disable|install|doctor
openclaw plugins install --force# v2026.4.5+: replace existing without dangerous override
openclaw plugins install clawhub:<pkg>  # ClawHub-first installs
openclaw skills list|info|check   # v2026.3.22+: search|install|update
clawhub install|uninstall <slug>
clawhub update --all

Sessions & Update CLI

openclaw sessions cleanup       # Disk budget cleanup
openclaw memory search --query "text"
openclaw memory rem-harness      # v2026.4.5: REM preview (experimental dreaming)
openclaw memory promote-explain  # v2026.4.5: explain promotion
openclaw cron list|add|edit|rm|enable|disable|run  # v2026.4.1: --tools on add|edit
openclaw flows list|show|cancel  # Task flows (substrate v2026.4.2+)
openclaw config schema         # JSON Schema dump (v2026.3.28+)
openclaw --container my-openclaw status --all --deep # CLI in container (v2026.3.24)
openclaw update --dry-run       # Preview update
openclaw update --channel stable|beta|dev
openclaw update --tag main  # From GitHub main (v2026.3.22+)

Backup CLI

openclaw backup create         # Local state archive
openclaw backup create --only-config # Config only
openclaw backup create --no-include-workspace
openclaw backup verify <path>    # Validate manifest
💬

Slash Commands

Slash commands are processed by the gateway before the AI sees them, making them instant and deterministic. Use them to reset sessions, control thinking levels, check token usage, or abort a runaway action. They work identically across Telegram, WhatsApp, Discord, and webchat.

In-Chat Commands Slash

Processed by Gateway — never reaches the AI.

CommandWhat it does
/statusModel, context, queue, runtime; adds provider usage when available
/tasksSession background task board (v2026.4.1)
/dreamingMemory dreaming status/help — experimental (v2026.4.5); see changelog
/help, /commandsCommon commands grouped
/new [model]Fresh session (alias: /reset)
/compact [instr]Summarize context, free tokens
/stopAbort run + clear queue
/think [level]off, minimal, low, medium, high, xhigh, adaptive
/model [id]Switch model; /model list, /model status
/usage off|tokens|full|costPer-response footer; cost = local summary
/fastToggle fast mode (Anthropic/OpenAI) — v2026.3.12
/tts on|off|always|inboundToggle text-to-speech
/send on|off|inheritOverride delivery (owner-only)
/reasoning on|off|streamSeparate reasoning message
/elevated on|off|ask|fullExec approval level
/whoamiSender identity (alias: /id)
/context [list|detail|json]Token usage per-file
/subagents spawn|list|kill|steerSpawn or control sub-agents
/export-session [path]Export session to HTML with full prompt
/approve allow-once|denyResolve exec approval prompts
/check-updatesQuick update summary

Directives Slash

:think high    # Thinking level
:model opus    # Model override
:fast         # Fast mode (Anthropic/OpenAI) v2026.3.12
:send off      # Suppress delivery
:verbose       # Verbose mode
:reasoning on  # Separate reasoning
:elevated ask  # Exec approval

Stop Phrases NEW

Natural language abort: “stop openclaw”, “stop action”, “please stop”, “STOP!”, “do not do that”. Works in ES/FR/ZH/HI/AR/JP/DE/PT/RU too.

📱

Channel Setup

Channel setup is where most users decide whether OpenClaw feels magical or frustrating. Telegram is the easiest stable path, while WhatsApp and Discord offer different tradeoffs around reliability, voice, and group workflows. Use the config blocks below as baseline templates before customizing policy and routing.

Telegram Config

{ channels: { telegram: {
  enabled: true,
  botToken: "123456:ABC-DEF..."
}}}

Most stable — survives reboots. DM voice transcription supported.

WhatsApp Config

{ channels: { whatsapp: {
  enabled: true,
  allowFrom: ["+1555..."],
  groupPolicy: "allowlist"
}}}

QR scan needed. Breaks on reboot.

Discord Config

{ channels: { discord: {
  enabled: true,
  token: "BOT_TOKEN",
  allowFrom: ["USER_ID"],
  autoArchiveDuration: 4320  // 1h=60, 1d=1440, 3d=4320, 1w=10080
}}}

Voice joins recover from DAVE failures. autoArchiveDuration for thread archiving (v2026.3.11).

All Channels

ChannelNotes
TelegramMost stable. Bot API. DM voice.
WhatsAppQR scan. Breaks on reboot.
DiscordVoice + text. DAVE recovery.
Google ChatNative support.
SlackBot + App Token + Signing Secret.
iMessagemacOS only, Shortcuts bridge.
SignalRequires signal-cli.
MS TeamsExtension channel.
Synology ChatNative plugin. Webhook ingress. NEW
WebChatBuilt-in at localhost:18789
MatrixExtension. BlueBubbles, Zalo also.

📱 Full Channels Guide NEW

Side-by-side comparison of every channel, setup configs for Telegram, WhatsApp, Discord, iMessage, Slack & Signal, plus multi-channel and troubleshooting tips.

📱 Open Channels Guide
🧠

Memory System

OpenClaw remembers what matters through a layered memory system: daily logs capture conversation history, MEMORY.md holds curated long-term facts, and vector search lets the agent recall context from weeks ago. Optional dreaming (v2026.4.5+) can promote short-term notes into durable recall in the background; when enabled, a top-level dreams.md holds the operator-visible trail while /dreaming and the Dreams UI surface status. Understanding flush, compaction, and indexing helps you stay sharp without bloating token usage.

How It Works

  • Daily logs memory/YYYY-MM-DD.md — today + yesterday
  • Long-term MEMORY.md — curated, DM only
  • Vector search — ~400 token chunks
  • Providers: local GGUF → OpenAI → Gemini (incl. gemini-embedding-2-preview) → Ollama → Voyage
  • Flush at 40k tokens → daily log
  • FTS: multilingual stop-words (ES/PT/JP/KR/AR) NEW

CLI CLI

openclaw memory index --all
openclaw memory search --query "text"
openclaw memory rem-harness      # REM preview (experimental dreaming, v2026.4.5+)
openclaw memory promote-explain  # Explain promotion decisions
openclaw sessions cleanup  # Disk budget
  • “Save this to memory” for important facts
  • Keep MEMORY.md compact
  • Doctor now checks memory embedding readiness

Dreaming v2026.4.5 · experimental

Dreaming runs weighted promotion from recent daily notes into longer-lived memory (light / deep / REM-style phases). Promotion trace and diary-style output are written to dreams.md at the workspace root so you can inspect what ran without pulling that file into default recall unless you ask.

  • Chat: /dreaming — status and help for the feature
  • UI: Dreams / Dream Diary in the Control UI when dreaming is enabled
  • Config: opt-in via dreaming settings (see upstream docs); tune aging with recencyHalfLifeDays / maxAgeDays where supported

Changelog: Memory dreaming → · Memory guide

🧩

Skills & ClawHub

Skills extend what your assistant can do — from web searches and calendar integration to custom automation scripts. Bundled skills ship with OpenClaw and require zero setup, while community skills on ClawHub let you install with one command. You can also build your own SKILL.md files for private tools and workflows.

SKILL.md File

---
name: my-skill
description: "What it does"
metadata: { "openclaw": { "requires": {
  "bins": ["uv"], "env": ["API_KEY"]
}}}
---

Frontmatter

KeyDefaultEffect
user-invocabletrue/slash command
disable-model-invocationfalseExclude from prompt
command-dispatch“tool” = bypass model
osalldarwin, linux, win32

Priority

  • Workspace (highest) — workspace/skills/
  • Managed~/.openclaw/skills/
  • Bundled (lowest)

🧩 Browse & Build Skills NEW

Explore 40+ curated skills by category, copy install commands, and generate custom SKILL.md files with the interactive builder.

🧩 Open Skill Explorer
💓

Heartbeat System

The heartbeat is your assistant's proactive loop — it runs on a schedule, reads HEARTBEAT.md, and only notifies you when something needs attention. Use it for server health checks, inbox monitoring, weather alerts, or any recurring task that should run silently until it matters.

How It Works

  • Default: every 30m (1h OAuth)
  • Reads HEARTBEAT.md strictly
  • HEARTBEAT_OK → suppressed
  • Empty → skipped
  • every: "0m" → disabled
  • Heartbeat deliveries now skip DM targets by default

Config Config

heartbeat: {
  every: "30m", target: "last",
  model: "cheap-model",
  activeHours: { start: "08:00", end: "22:00",
    timezone: "Europe/Bucharest" }
}

Cron Jobs

Cron jobs let your assistant run scheduled tasks independently — morning briefings, weekly reports, CI checks, or data scraping. Jobs are defined in cron/jobs.json with standard cron expressions and can target isolated sessions so they never interfere with your active conversations.

Job Format Config

{ "name": "Morning Brief",
  "schedule": { "kind": "cron", "expr": "0 8 * * *" },
  "sessionTarget": "isolated",
  "payload": { "kind": "agentTurn",
    "message": "Deliver my briefing." }
}

Schedules Tip

  • 0 7 * * * — Morning briefing
  • 0 18 * * 1-5 — Weekday recap
  • 0 10 * * 1 — Weekly intel
  • 0 8 1 * * — Monthly reminder

Cron persists delivered state for failure visibility. v2026.3.11: Isolated cron jobs no longer notify via ad hoc agent sends — run openclaw doctor --fix to migrate legacy config.

🤖

Model Providers

Model selection drives quality, speed, and cost. Use a high-quality model for deep reasoning and a cheaper model for heartbeat or routine automation. The matrix below is designed to help you pick provider formats, auth patterns, and fallback strategy without burning budget.

Supported

ProviderFormatAuth
Anthropicanthropic/claude-*API / OAuth
OpenAIopenai/gpt-*API / OAuth
Mistral NEWmistral/*API key
Googlegoogle/gemini-*, gemini-3.1-flash-liteAPI key
OpenRouteropenrouter/*Single key
Kilo Gateway NEWkilocode/*Auth + onboarding
Vercel AI GWvercel-ai-gateway/*Normalized refs
Amazon Bedrockamazon-bedrock/*AWS SDK
Ollamaollama/*Local
LM StudioCustom baseUrlLocal
Moonshot NEWkimi providerAPI key
Doubao/BytePlus NEWvolcengine providervolcengine-api-key

Local LLM Config

models: { providers: { "ollama": {
  baseUrl: "http://127.0.0.1:11434/v1",
  apiKey: "ollama",
  api: "openai-responses"
}}}

Cost Tips Tip

  • Cheap model for heartbeats
  • Claude Max $200/mo = no per-token
  • Gemini free tier for basics
  • Ollama = $0, local, private
  • cacheRetention per-agent for prompt caching
  • Bedrock/OpenRouter cache behavior documented
  • v2026.3.1+: Claude 4.6 thinking defaults to adaptive; other reasoning models default to low

💰 Calculate Your Real Costs NEW

Interactive calculator with all providers, heartbeat costs, optimizations, and savings tips. Know exactly what you'll spend.

💰 Open Cost Calculator
🔐

Security Essentials

OpenClaw can read files, run commands, and connect to external services, so security defaults matter from day one. Start with DM pairing, allowlists, and sandboxing, then add skill vetting and regular audits. Treat every inbound message and community skill as untrusted input until proven safe.

Hardening Security

openclaw doctor --fix
openclaw security audit --deep
chmod 700 ~/.openclaw
chmod 600 ~/.openclaw/openclaw.json
chmod 700 ~/.openclaw/credentials

Policies Config

security: {
  dmPolicy: "pairing",
  groupPolicy: "allowlist",
  sandboxEnabled: true,
  trust_model: {
    multi_user_heuristic: true  // NEW
  }
}
  • Browser SSRF now trusted-network by default
  • Exec obfuscation detection before allowlist
  • HSTS for direct HTTPS deployments

Skill Vetting Security

  • Red flags: curl|bash, base64, Gatekeeper bypass
  • pip install cisco-ai-skill-scanner
  • Check zero-width Unicode in SKILL.md
  • No auto-update skills
  • Sandbox rejects hardlink alias escapes
  • v2026.3.12: Workspace plugin auto-load disabled — explicit trust required for workspace plugins

Owner-Only Commands v2026.3.12

/config and /debug now require sender ownership. Authorized non-owner senders can no longer reach these surfaces.

🔐 Full Security & DM Policies Guide NEW

Deep dive: DM policies, pairing flow, sandboxing, exec controls, credential hygiene, network hardening, skill vetting, incident response & a 20-point checklist.

🔐 Open Security Guide
🔧

Troubleshooting

Most OpenClaw issues fall into a few categories: port conflicts, stale sessions, channel disconnects, or configuration mismatches. The openclaw doctor --deep --yes command resolves the majority automatically. This section covers the rest — common problems, their fixes, and the update commands to keep your installation current.

Looking for a specific error? → Browse all 20 error fixes

Universal Fix

openclaw doctor --deep --yes

Solves 80% of issues. Now checks memory embedding readiness too.

Common Issues

ProblemFix
Port conflictlsof -i :18789
Context full/compact or /new
Channel downopenclaw channels status --probe
No responseopenclaw logs --follow
WhatsApp dropsRe-scan QR. Use Telegram.
Stale lockAuto-detected via port reachability now
Config invalid / startup failsopenclaw config validate for JSON/schema errors
Auth/pairing fails after upgradeSet gateway.auth.mode to token or password (v2026.3.7+). Cron notify broken? Run openclaw doctor --fix (v2026.3.11). Workspace plugins not loading? v2026.3.12 disabled auto-load — explicit trust required.
Restart loopFixed in v2026.2.22+. Update.
Disk full sessionsopenclaw sessions cleanup

Update CLI

sudo npm i -g openclaw@latest
openclaw update --dry-run
openclaw update wizard
openclaw update --channel beta
📖

Quick Glossary

OpenClaw uses specific terminology throughout its docs and config. If a term in another section is unclear, check here first. These definitions are kept short and practical — they map directly to concepts you will encounter during setup and daily use.

Gateway
Background process routing messages between channels and the AI. Runs via launchd/systemd.
Workspace
Agent's home (~/.openclaw/workspace/). All .md files defining behavior, memory, personality.
SOUL.md
Agent personality, tone, values. The "character sheet." Loaded every session.
ClawHub
Community skill registry. clawhub install <slug>.
Heartbeat
Periodic check (default 30m). Reads HEARTBEAT.md, notifies only when needed.
Compaction
Summarizes long context → daily log. Auto at threshold or /compact.
Dreaming
Experimental background memory promotion (v2026.4.5+). dreams.md trail; /dreaming; Dreams UI.
Pairing
Security linking of user identity to channel. Prevents unauthorized access.
Sandbox
Isolated exec environment for skills. Rejects hardlink escapes.
Subagent
Spawnable child agent via /subagents spawn. Deterministic activation. NEW
Canvas
Live UI surface the agent can push/reset. A2UI pattern.
JSON5
Config format. Comments (//) and trailing commas OK.
Kilo Gateway
New provider routing through kilocode. Default: claude-opus-4.6. NEW
📂

Key Paths

Knowing where OpenClaw stores its files is essential for backups, debugging, and manual edits. The workspace lives in ~/.openclaw/workspace/, config in ~/.openclaw/openclaw.json, and credentials in a permissions-locked directory. Gateway logs rotate daily under /tmp/openclaw/.

Directory Structure

~/.openclaw/
├── openclaw.json           ← Main config (JSON5)
├── credentials/             ← OAuth/API keys (chmod 600)
├── workspace/               ← Agent's home
│   ├── AGENTS.md  SOUL.md  USER.md  IDENTITY.md
│   ├── TOOLS.md  HEARTBEAT.md  MEMORY.md  dreams.md
│   ├── memory/              ← Daily logs
│   └── skills/              ← Workspace skills
├── agents/<id>/sessions/    ← Chat history
├── skills/                  ← Managed skills
├── memory/<id>.sqlite       ← Vector index
├── cron/jobs.json           ← Scheduled tasks
└── sandboxes/               ← Sandbox workspaces

/tmp/openclaw/openclaw-YYYY-MM-DD.log  ← Gateway log
🔊

Voice & TTS

Voice and text-to-speech turn OpenClaw into a hands-free assistant. Choose from premium providers like ElevenLabs, fast options like OpenAI TTS, or free alternatives like Edge TTS. Voice input works natively in Telegram and Discord, and the browser canvas supports real-time interaction with snapshot and action controls.

Providers

ProviderQualityCost
ElevenLabsUltra-realisticPremium
OpenAI TTSFast, high-qualityStandard
Edge TTSMulti-languageFree
Mistral VoiceNew provider NEWStandard

Talk config is now provider-agnostic. iOS TTS prefetches segments for smooth playback. macOS voice wake defaults to webchat channel.

Browser & Canvas CLI

openclaw browser fill --fields '[...]'
openclaw sandbox list|recreate

Canvas: A2UI push/reset/eval/snapshot. Browser: dedicated Chrome profiles, snapshots, actions, uploads.