OpenClaw
Web: https://openclaw.ai/
2026.4.15
# OpenClaw/2026.4.15
module load ceuadmin/node/24.15.0
export version=2026.4.15
export BASE="$CEUADMIN/OpenClaw/$version"
npm view openclaw versions
npm install -g "openclaw@$version" --prefix "$BASE"
# Extensions
module load ceuadmin/OpenClaw/$version
export OPENCLAW_STATE_DIR="$BASE/state"
export OPENCLAW_CONFIG_PATH="$BASE/config.json"
openclaw plugins install ollama
from which one finds
- $BASE/lib/node_modules/openclaw/dist/plugin-sdk/extensions/ollama
- $BASE/lib/node_modules/openclaw/dist/extensions/ollama
It turns out node/22.16.0 is a bit too old, so we resort to 24.15.0 (LTS).
npm warn EBADENGINE Unsupported engine {
npm warn EBADENGINE package: 'undici@8.0.2',
npm warn EBADENGINE required: { node: '>=22.19.0' },
npm warn EBADENGINE current: { node: 'v22.16.0', npm: '10.9.2' }
npm warn EBADENGINE }
npm warn deprecated inflight@1.0.6: This module is not supported, and leaks memory. Do not use it. Check out lru-cache if you want a good and tested way to coalesce async requests by a key value, which is much more comprehensive and powerful.
npm warn deprecated npmlog@5.0.1: This package is no longer supported.
npm warn deprecated rimraf@3.0.2: Rimraf versions prior to v4 are no longer supported
npm warn deprecated are-we-there-yet@2.0.0: This package is no longer supported.
npm warn deprecated glob@7.2.3: Old versions of glob are not supported, and contain widely publicized security vulnerabilities, which have been fixed in the current version. Please update. Support for old versions may be purchased (at exorbitant rates) by contacting i@izs.me
npm warn deprecated gauge@3.0.2: This package is no longer supported.
npm warn deprecated node-domexception@1.0.0: Use your platform's native DOMException instead
npm warn deprecated tar@6.2.1: Old versions of tar are not supported, and contain widely publicized security vulnerabilities, which have been fixed in the current version. Please update. Support for old versions may be purchased (at exorbitant rates) by contacting i@izs.me
Setup
It has a flavour of settings for libvips.
source ~/rds/software/py3.11/bin/activate
module load ceuadmin/libvips/8.18.0
module load ceuadmin/node/22.16.0
npm install -g node-gyp --prefix "$CEUADMIN/OpenClaw/2026.3.8"
npm install -g openclaw --prefix "$CEUADMIN/OpenClaw/2026.3.8"
so that the openclaw binary and packages are installed according to –prefix.
Alternatively,
NPM_CONFIG_PREFIX=$CEUADMIN/OpenClaw/2026.3.8 curl -fsSL https://openclaw.ai/install.sh | bash
# npm install -g openclaw --build-from-source --prefix "$CEUADMIN/OpenClaw/2026.3.8"
OpenClaw is effectively installed as a node modules sitting with node/22.16.0 as a matter of convenience.
Usage
module load ceuadmin/OpenClaw
openclaw --help
which gives,
OpenClaw 2026.3.8 (3caab92)
🦞 OpenClaw 2026.3.8 (3caab92) — I'm the middleware between your ambition and your attention span.
Usage: openclaw [options] [command]
Options:
--dev Dev profile: isolate state under ~/.openclaw-dev, default gateway port 19001, and shift derived ports (browser/canvas)
-h, --help Display help for command
--log-level <level> Global log level override for file + console (silent|fatal|error|warn|info|debug|trace)
--no-color Disable ANSI colors
--profile <name> Use a named profile (isolates OPENCLAW_STATE_DIR/OPENCLAW_CONFIG_PATH under ~/.openclaw-<name>)
-V, --version output the version number
Commands:
Hint: commands suffixed with * have subcommands. Run <command> --help for details.
acp * Agent Control Protocol tools
agent Run one agent turn via the Gateway
agents * Manage isolated agents (workspaces, auth, routing)
approvals * Manage exec approvals (gateway or node host)
backup * Create and verify local backup archives for OpenClaw state
browser * Manage OpenClaw's dedicated browser (Chrome/Chromium)
channels * Manage connected chat channels (Telegram, Discord, etc.)
clawbot * Legacy clawbot command aliases
completion Generate shell completion script
config * Non-interactive config helpers (get/set/unset/file/validate). Default: starts setup wizard.
configure Interactive setup wizard for credentials, channels, gateway, and agent defaults
cron * Manage cron jobs via the Gateway scheduler
daemon * Gateway service (legacy alias)
dashboard Open the Control UI with your current token
devices * Device pairing + token management
directory * Lookup contact and group IDs (self, peers, groups) for supported chat channels
dns * DNS helpers for wide-area discovery (Tailscale + CoreDNS)
docs Search the live OpenClaw docs
doctor Health checks + quick fixes for the gateway and channels
gateway * Run, inspect, and query the WebSocket Gateway
health Fetch health from the running gateway
help Display help for command
hooks * Manage internal agent hooks
logs Tail gateway file logs via RPC
memory * Search and reindex memory files
message * Send, read, and manage messages
models * Discover, scan, and configure models
node * Run and manage the headless node host service
nodes * Manage gateway-owned node pairing and node commands
onboard Interactive onboarding wizard for gateway, workspace, and skills
pairing * Secure DM pairing (approve inbound requests)
plugins * Manage OpenClaw plugins and extensions
qr Generate iOS pairing QR/setup code
reset Reset local config/state (keeps the CLI installed)
sandbox * Manage sandbox containers for agent isolation
secrets * Secrets runtime reload controls
security * Security tools and local config audits
sessions * List stored conversation sessions
setup Initialize local config and agent workspace
skills * List and inspect available skills
status Show channel health and recent session recipients
system * System events, heartbeat, and presence
tui Open a terminal UI connected to the Gateway
uninstall Uninstall the gateway service + local data (CLI remains)
update * Update OpenClaw and inspect update channel status
webhooks * Webhook helpers and integrations
Examples:
openclaw models --help
Show detailed help for the models command.
openclaw channels login --verbose
Link personal WhatsApp Web and show QR + connection logs.
openclaw message send --target +15555550123 --message "Hi" --json
Send via your web session and print JSON result.
openclaw gateway --port 18789
Run the WebSocket Gateway locally.
openclaw --dev gateway
Run a dev Gateway (isolated state/config) on ws://127.0.0.1:19001.
openclaw gateway --force
Kill anything bound to the default gateway port, then start it.
openclaw gateway ...
Gateway control via WebSocket.
openclaw agent --to +15555550123 --message "Run summary" --deliver
Talk directly to the agent using the Gateway; optionally send the WhatsApp reply.
openclaw message send --channel telegram --target @mychat --message "Hi"
Send via your Telegram bot.
Docs: docs.openclaw.ai/cli
Integration with Ollama
URL, https://docs.ollama.com/integrations/openclaw
The official news on OpenClaw with ollama involves ollama launch openclaw where OpenClaw tries to install a systemd service, which
is not allowed on HPC login nodes, so we start manually.
module load ceuadmin/ollama
ollama serve > /dev/null 2>&1 &
until ollama list; do
sleep 1
done
module load ceuadmin/OpenClaw
openclaw onboard
openclaw gateway &
where openclaw onboard sets up the environment (from ollama 0.18, it is official provider for OpenClaw, i.e., openclaw onboard --auth-choice ollama) and our instance of openclaw gateway gives,
🦞 OpenClaw 2026.3.8 (3caab92) — Your personal assistant, minus the passive-aggressive calendar reminders.
15:53:10 [canvas] host mounted at http://127.0.0.1:18789/__openclaw__/canvas/ (root /home/jhz22/.openclaw/canvas)
15:53:10 [heartbeat] started
15:53:10 [health-monitor] started (interval: 300s, startup-grace: 60s, channel-connect-grace: 120s)
15:53:10 [gateway] agent model: ollama/minimax-m2.5:cloud
15:53:10 [gateway] listening on ws://127.0.0.1:18789 (PID 322663)
15:53:10 [gateway] log file: /tmp/openclaw/openclaw-2026-03-11.log
15:53:10 [browser/server] Browser control listening on http://127.0.0.1:18791/ (auth=token)
Surely, these are useful for letting web browsers point to the appropriate ports.
Now we try ollama launch openclaw again,
This will modify your OpenClaw configuration:
/home/jhz22/.openclaw/openclaw.json
Backups will be saved to /rds/user/jhz22/hpc-work/work/ollama-backups/
Added minimax-m2.5:cloud to OpenClaw
Launching OpenClaw with minimax-m2.5:cloud...
Security
OpenClaw can read files and run actions when tools are enabled.
A bad prompt can trick it into doing unsafe things.
Learn more: https://docs.openclaw.ai/gateway/security
I understand the risks. Continue?
Yes No
✓ Installed web search plugin
Preparing your assistant — this may take a moment...
17:54:21 [reload] config change detected; evaluating reload (tools.web, plugins)
17:54:21 [reload] config change requires gateway restart (plugins)
17:54:21 [gateway] signal SIGUSR1 received
17:54:21 [gateway] received SIGUSR1; restarting
17:54:56 [gateway] shutdown timed out; exiting without full cleanup
Warning: gateway did not come back after restart
Starting gateway...
✓ OpenClaw is running
Open the Web UI:
http://localhost:18789/#token=ollama
Quick start:
/help see all commands
openclaw configure --section channels connect WhatsApp, Telegram, etc.
openclaw skills browse and install skills
The OpenClaw gateway is running in the background.
Stop it with: openclaw gateway stop
🦞 OpenClaw 2026.3.8 (3caab92) — Half butler, half debugger, full crustacean.
openclaw tui - ws://127.0.0.1:18789 - agent main - session main
session agent:main:main
Wake up, my friend!
Morning! ☀️ I'm here. Nothing on the agenda per HEARTBEAT.md — what can I do for you?
how are you?
Doing well, thanks for asking! Freshly woken up and ready to go. What's up?
connected | idle
agent main | session main (openclaw-tui) | ollama/minimax-m2.5:cloud | think low | tokens 9.3k/205k (5%)
Note that additionally, we have
# connect WhatsApp, Telegram, etc.
openclaw configure --section channels
# browse and install skills
openclaw skills
# start a specific model
ollama launch openclaw --model kimi-k2.5:cloud
ollama launch openclaw --model gemma4:31b-cloud
ollama launch openclaw --model glm-5.1:cloud
The web UI http://localhost:18789/#token=ollama (or http://localhost:18789/chat?session=main) actually provides a gateway dashboard shown below, which is more desirable.
MiniMax-M2.7
This model is designed to excel at completing end-to-end tasks in software engineering and office productivity while improving its character and emotional intelligence.
M2.7 is also good at self-improvement using data from the web. Start the model with ollama launch openclaw --model minimax-m2.7:cloud, or when OpenClaw is running, /model ollama/minimax-m2.7:cloud. An exemplar use of self-improvement and research can be done with
Teach yourself how to write financial research reports by studying examples from top analysts. Write a PPT presentation and Word document report on TSMC based on their latest earnings. Save your research framework as a skill so future reports follow the same process.
To launch an agent non-interactively (–yes),
ollama launch openclaw \
--model minimax-m2.7:cloud \
--yes -- agent \
--agent main \
--local \
--message "Prepare a pre-read for my next meeting"
ClawHub
Web, https://clawhub.ai/
Some notable skills are liisted below,
module load ceuadmin/node/24.15.0
npx clawhub@latest install self-improving-agent
npx clawhub@latest install tavily-search
npx clawhub@latest install gog
npx clawhub@latest install github
npx clawhub@latest install summarize
npx clawhub@latest install agent-browser
npx clawhub@latest install find-skills
Somehow the installation is destined to skills/skills, nevertheless can be reset manually by mving skills to the right place, e.g.,
$ npx clawhub@latest install self-improving-agent
✔ OK. Installed self-improving-agent -> /usr/local/Cluster-Apps/ceuadmin/OpenClaw/2026.4.15/lib/node_modules/openclaw/skills/skills/self-improving-agent
16:58 jhz22@login-q-1 /usr/local/Cluster-Apps/ceuadmin/OpenClaw/2026.4.15/lib/node_modules/openclaw/skills $ openclaw skills list
🦞 OpenClaw 2026.4.15 (041266a) — I've survived more breaking changes than your last three relationships.
Skills (1/1 ready)
┌──────────┬──────────────────────┬──────────────────────────────────────────────────────────────────────────────────────────┬──────────────────┐
│ Status │ Skill │ Description │ Source │
├──────────┼──────────────────────┼──────────────────────────────────────────────────────────────────────────────────────────┼──────────────────┤
│ ✓ ready │ 📦 self-improvement │ Captures learnings, errors, and corrections to enable continuous improvement. Use when: │ openclaw-bundled │
│ │ │ (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's │ │
│ │ │ wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An │ │
│ │ │ external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, │ │
│ │ │ (6) A better approach is discovered for a recurring task. Also review learnings before │ │
│ │ │ major tasks. │ │
└──────────┴──────────────────────┴──────────────────────────────────────────────────────────────────────────────────────────┴──────────────────┘
Tip: use `openclaw skills search`, `openclaw skills install`, and `openclaw skills update` for ClawHub-backed skills.
$ cd /usr/local/Cluster-Apps/ceuadmin/OpenClaw/2026.4.15/lib/node_modules/openclaw/skills
$ mv skills/self-improving-agent/ .
$ rmdir skills
$ openclaw skills list
$ cd -