shinejh0528/openclaw

Openclaw

Supported -e Environment Variables (entrypoint.sh)

• OPENCLAW_GATEWAY_PORT : Gateway port inside the container. Default: ***

  • Example: -e OPENCLAW_GATEWAY_PORT=18789

• OPENCLAW_GATEWAY_TOKEN : Gateway auth token. If provided, it is used as-is. If not provided, the entrypoint generates one via openssl rand -hex 32, then saves it to /root/.openclaw/gateway.token and reuses it on later runs.

  • Example: -e OPENCLAW_GATEWAY_TOKEN=xxxxxxxx

• LLM_PLATFORM : LLM provider selector. In the current entrypoint, only openai is handled explicitly (it writes OPENAI_API_KEY internally).

  • Allowed (major):
    • openai (OpenAI API + Codex)
    • anthropic (Claude API)
    • openrouter (OpenRouter)
    • litellm (LiteLLM unified gateway)
    • bedrock (Amazon Bedrock)
    • vercel (Vercel AI Gateway)
    • together (Together AI)
    • cloudflare (Cloudflare AI Gateway)
    • moonshot (Moonshot AI / Kimi)
    • venice (Venice AI)
    • ollama (local models)
    • qwen (Qwen OAuth)
  • Example: -e LLM_PLATFORM=openai

• LLM_MODEL : Model name.

  • If you pass gpt-5-mini, it is normalized to openai/gpt-5-mini.
  • If you pass openai/<model>, it is used as-is.
  • Default: gpt-5-mini (effective: openai/gpt-5-mini)
  • Example: -e LLM_MODEL=gpt-5-mini

• LLM_API_KEY : API key for the selected provider. When LLM_PLATFORM=openai, it is written to OPENAI_API_KEY in /root/.openclaw/.env.

  • Example: -e LLM_API_KEY=sk-...

• BRAVE_SEARCH_ENABLED : Enables Brave Search key injection (env-only).

  • Allowed: true / false
  • Default: false
  • Example: -e BRAVE_SEARCH_ENABLED=true

• BRAVE_API_KEY : Brave Search API key. Used only when BRAVE_SEARCH_ENABLED=true. Stored in /root/.openclaw/.env.

  • Example: -e BRAVE_API_KEY=...

• CHAT_PLATFORM : Chat platform selector. In the current entrypoint, this is only used to decide whether to store the *** token (no pairing automation).

  • Allowed: telegram (others are ignored)
  • Example: -e CHAT_PLATFORM=telegram

• ***_BOT_TOKEN : *** bot token. Stored in /root/.openclaw/.env only when CHAT_PLATFORM=telegram. (Pairing/enable steps are not automated.)

  • Example: -e TELEGRAM_BOT_TOKEN=123456:ABC...

Usage

1. Run container

You can run with -itd option. In this example, we use -it option to see log.

PowerShell ex)

powershell
# =========================
# env variables
# =========================
$sOpenai_api_key = "sk-...YOUR_OPENAI_KEY..."
$sBrave_api_key = "BSA...YOUR_BRAVE_KEY..."
$sTelegram_bot_token = "123456789:ABC...YOUR_TELEGRAM_BOT_TOKEN..."

# =========================
# command
# =========================
docker run -it --name openclaw `
  -p 18789:18789 `
  -v F:\docker_volume\openclaw\.openclaw:/root/.openclaw `
  -e OPENCLAW_GATEWAY_PORT=18789 `
  -e LLM_PLATFORM=openai `
  -e LLM_MODEL=gpt-5-mini `
  -e LLM_API_KEY=$sOpenai_api_key `
  -e BRAVE_SEARCH_ENABLED=true `
  -e BRAVE_API_KEY=$sBrave_api_key `
  -e CHAT_PLATFORM=telegram `
  -e TELEGRAM_BOT_TOKEN=$sTelegram_bot_token `
  shinejh0528/openclaw:1.0.0

Linux cmd ex)

bash
# =========================
# env variables
# =========================
export OPENAI_API_KEY="sk-...YOUR_OPENAI_KEY..."
export BRAVE_API_KEY="BSA...YOUR_BRAVE_KEY..."
export TELEGRAM_BOT_TOKEN="123456789:ABC...YOUR_TELEGRAM_BOT_TOKEN..."

# =========================
# command
# =========================
docker run -it --name openclaw \
  -p 18789:18789 \
  -v /opt/docker_volume/openclaw/.openclaw:/root/.openclaw \
  -e OPENCLAW_GATEWAY_PORT=18789 \
  -e LLM_PLATFORM=openai \
  -e LLM_MODEL=gpt-5-mini \
  -e LLM_API_KEY="${OPENAI_API_KEY}" \
  -e BRAVE_SEARCH_ENABLED=true \
  -e BRAVE_API_KEY="${BRAVE_API_KEY}" \
  -e CHAT_PLATFORM=telegram \
  -e TELEGRAM_BOT_TOKEN="${TELEGRAM_BOT_TOKEN}" \
  shinejh0528/openclaw:1.0.0

2. Enter inside the container and onboard

docker exec -it openclaw /bin/bash
openclaw onboard

Setting for onboard

🦞 OpenClaw 2026.2.9 (<commit>) — <tagline>

▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
██░▄▄▄░██░▄▄░██░▄▄▄██░▀██░██░▄▄▀██░████░▄▄▀██░███░██
██░███░██░▀▀░██░▄▄▄██░█░█░██░█████░████░▀▀░██░█░█░██
██░▀▀▀░██░█████░▀▀▀██░██▄░██░▀▀▄██░▀▀░█░██░██▄▀▄▀▄██
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
                  🦞 OPENCLAW 🦞

┌  OpenClaw onboarding
│
◇  Security ──────────────────────────────────────────────────────────────────────────────╮
│                                                                                         │
│  Security warning — please read.                                                        │
│                                                                                         │
│  OpenClaw is a hobby project and still in beta. Expect sharp edges.                     │
│  This bot can read files and run actions if tools are enabled.                          │
│  A bad prompt can trick it into doing unsafe things.                                    │
│                                                                                         │
│  If you’re not comfortable with basic security and access control, don’t run OpenClaw.  │
│  Ask someone experienced to help before enabling tools or exposing it to the internet.  │
│                                                                                         │
│  Recommended baseline:                                                                  │
│  - Pairing/allowlists + mention gating.                                                 │
│  - Sandbox + least-privilege tools.                                                     │
│  - Keep secrets out of the agent’s reachable filesystem.                                │
│  - Use the strongest available model for any bot with tools or untrusted inboxes.       │
│                                                                                         │
│  Run regularly:                                                                         │
│  openclaw security audit --deep                                                         │
│  openclaw security audit --fix                                                          │
│                                                                                         │
│  Must read: https://docs.openclaw.ai/gateway/security                                   │
│                                                                                         │
├─────────────────────────────────────────────────────────────────────────────────────────╯
│
◇  I understand this is powerful and inherently risky. Continue?
│  Yes
│
◇  Onboarding mode
│  QuickStart
│
◇  Existing config detected ─────────╮
│                                    │
│  workspace: ~/.openclaw/workspace  │
│  model: openai/gpt-5-mini          │
│  gateway.mode: local               │
│  gateway.port: 18789               │
│  gateway.bind: loopback            │
│  skills.nodeManager: pnpm          │
│                                    │
├────────────────────────────────────╯
│
◇  Config handling
│  Use existing values
│
◇  QuickStart ─────────────────────────────╮
│                                          │
│  Keeping your current gateway settings:  │
│  Gateway port: 18789                     │
│  Gateway bind: Loopback (127.0.0.1)      │
│  Gateway auth: Token (default)           │
│  Tailscale exposure: Off                 │
│  Direct to chat channels.                │
│                                          │
├──────────────────────────────────────────╯
│
◇  Model/auth provider
│  OpenAI
│
◇  OpenAI auth method
│  OpenAI API key
│
◇  Use existing OPENAI_API_KEY (env: OPENAI_API_KEY, sk-…<redacted>)?
│  Yes
│
◇  OpenAI API key ──────────────────────────────────────────────────────────╮
│                                                                           │
│  Saved OPENAI_API_KEY to /root/.openclaw/.env for launchd compatibility.  │
│                                                                           │
├───────────────────────────────────────────────────────────────────────────╯
│
◇  Model configured ──────────────────────────╮
│                                             │
│  Default model set to openai/<redacted>     │
│                                             │
├─────────────────────────────────────────────╯
│
◇  Default model
│  openai/gpt-5-mini
│
◇  Channel status ────────────────────────────╮
│                                             │
│  Telegram: configured                       │
│  WhatsApp: not configured                   │
│  Discord: not configured                    │
│  Google Chat: not configured                │
│  Slack: not configured                      │
│  Signal: not configured                     │
│  iMessage: work in progress                 │
│  (Other channels omitted)                   │
│                                             │
├─────────────────────────────────────────────╯
│
◇  Select channel (QuickStart)
│  Telegram (Bot API)
│
◇  Telegram already configured. What do you want to do?
│  Skip (leave as-is)
Updated ~/.openclaw/openclaw.json
Workspace OK: ~/.openclaw/workspace
Sessions OK: ~/.openclaw/agents/main/sessions
│
◇  Skills status ────────────╮
│                            │
│  Eligible: <n>             │
│  Missing requirements: <n> │
│  Blocked by allowlist: <n> │
│                            │
├────────────────────────────╯
│
◇  Configure skills now? (recommended)
│  Yes
│
◇  Homebrew recommended ──────────────────────────────────────────────────────────╮
│                                                                                 │
│  Many skill dependencies are shipped via Homebrew.                              │
│  Without brew, you'll need to build from source or download releases manually.  │
│                                                                                 │
├─────────────────────────────────────────────────────────────────────────────────╯
│
◇  Hooks ──────────────────────────────────────────────────────────╮
│                                                                  │
│  Hooks let you automate actions when agent commands are issued.  │
│  Example: Save session context to memory when you issue /new.    │
│                                                                  │
│  Learn more: https://docs.openclaw.ai/hooks                      │
│                                                                  │
├──────────────────────────────────────────────────────────────────╯
│
◇  Enable hooks?
│  📝 command-logger
│
◇  Control UI ─────────────────────────────────────────────────────────────────────╮
│                                                                                  │
│  Web UI: http://127.0.0.1:18789/                                                 │
│  Web UI (with token): http://127.0.0.1:18789/#token=<REDACTED_TOKEN>             │
│  Gateway WS: ws://127.0.0.1:18789                                                │
│  Docs: https://docs.openclaw.ai/web/control-ui                                   │
│                                                                                  │
├──────────────────────────────────────────────────────────────────────────────────╯
│
◇  Token ─────────────────────────────────────────────────────────────────────────────────╮
│                                                                                         │
│  Gateway token: shared auth for the Gateway + Control UI.                               │
│  Stored in: ~/.openclaw/openclaw.json (gateway.auth.token) or OPENCLAW_GATEWAY_TOKEN.   │
│  View token: openclaw config get gateway.auth.token                                     │
│  Generate token: openclaw doctor --generate-gateway-token                               │
│  Web UI stores a copy in this browser's localStorage (openclaw.control.settings.v1).    │
│  If prompted: paste the token into Control UI settings (or use the tokenized URL).     │
│                                                                                         │
├─────────────────────────────────────────────────────────────────────────────────────────╯
│
◇  Start TUI (best option!)
│  openclaw tui - ws://127.0.0.1:18789 - agent main - session main
│
│  Wake up, my friend!
│
│  Hey — I’m awake.
│  (Intro questions omitted)

3. Get *** pairing token and enroll the token

!https://raw.githubusercontent.com/Shin-jongwhan/docker/main/release_note/shinejh0528/openclaw/***_pairing_token.png

Cmd ex)

openclaw pairing approve telegram [your_token]

Run manually

docker run -itd -p 18789:18789 --name openclaw shinejh0528/openclaw:1.0.0 /bin/bash

# and enter inside the container, run openclaw manually
docker exec -it openclaw

Running with your settings

After the initial setup (mapping the .openclaw folder with the -v option), you can just run the command, and OpenClaw will start up right away.

docker run -it --name openclaw -p 18789:18789 -v F:\docker_volume\openclaw\.openclaw:/root/.openclaw shinejh0528/openclaw:1.0.0 openclaw gateway --port 18789 --verbose