1. Verify OpenClaw Installation

First ensure OpenClaw is properly installed and accessible.

openclaw --version

or:

which openclaw

Expected result: /usr/local/bin/openclaw

2. Recommended Ubuntu Environment

Component	Recommended
Ubuntu	22.04 LTS / 24.04 LTS
Python	3.10+
NodeJS	20+
RAM	Minimum 8 GB
CPU	4 cores
GPU	Optional but recommended
Docker	Optional

3. Create Working Directory

mkdir -p ~/openclaw-workspace
cd ~/openclaw-workspace

Recommended structure:

~/openclaw-workspace/
├── projects/
├── uploads/
├── outputs/
├── logs/
├── temp/
├── scripts/
└── models/

4. Configure OpenClaw

Main configuration usually uses openclaw.json. Create configuration:

nano openclaw.json

Example configuration:

{
 "workspace": "/home/ubuntu/openclaw-workspace",
 "memory": { "provider": "sqlite" },
 "llm": { "provider": "openai", "model": "gpt-4.1" },
 "logging": { "level": "info" },
 "ocr": { "enabled": true, "engine": "tesseract" }
}

5. Configure Environment Variables

nano ~/.bashrc

Add:

export OPENAI_API_KEY="YOUR_API_KEY"
export OPENCLAW_HOME="$HOME/openclaw-workspace"
export TESSDATA_PREFIX="/usr/share/tesseract-ocr/5/tessdata"

Reload: source ~/.bashrc

6. Install OCR Dependencies (Important)

sudo apt update
sudo apt install -y tesseract-ocr libtesseract-dev tesseract-ocr-eng tesseract-ocr-ind tesseract-ocr-mya imagemagick poppler-utils ffmpeg

Verify: tesseract --version
List languages: tesseract --list-langs
Expected: eng ind mya

7. Configure Myanmar OCR

sudo apt install tesseract-ocr-mya

Test OCR: tesseract sample-myanmar.png stdout -l mya
Mixed languages: tesseract sample.png stdout -l eng+ind+mya

8. Improve OCR Accuracy

Recommended preprocessing:

• Convert to grayscale: convert input.png -colorspace Gray output.png
• Increase contrast: convert input.png -brightness-contrast 20x40 output.png
• Resize: convert input.png -resize 300% output.png
• Denoise: convert input.png -despeckle output.png

9. Enable OpenClaw OCR Skill

openclaw plugins install tesseract-ocr

or openclaw skills install tesseract-ocr
Verify: openclaw plugins list

10. Running OpenClaw

Interactive Mode: openclaw
Project Mode: openclaw ./projects/myproject
Agent Mode: openclaw agent start
Debug Mode: openclaw --debug

11. Using OpenClaw with OCR

Example workflow: image upload → OCR extraction → LLM analysis → automation.
Example: openclaw analyze invoice.png
Outputs: text extraction, invoice analysis, translation, JSON generation, etc.

12. Telegram / WhatsApp Integration

Example Telegram configuration:

{ "telegram": { "enabled": true, "bot_token": "YOUR_BOT_TOKEN" } }

Start agent: openclaw telegram start
⚠️ OCR quality depends on image compression. Prefer file uploads and high-res screenshots.

13. Running OpenClaw as a Service

sudo nano /etc/systemd/system/openclaw.service

Content:

[Unit]
Description=OpenClaw Service
After=network.target

[Service]
Type=simple
User=ubuntu
WorkingDirectory=/home/ubuntu/openclaw-workspace
ExecStart=/usr/local/bin/openclaw
Restart=always

[Install]
WantedBy=multi-user.target

sudo systemctl daemon-reload
sudo systemctl enable openclaw
sudo systemctl start openclaw
sudo systemctl status openclaw

14. Logging and Monitoring

journalctl -u openclaw -f

or tail -f ~/openclaw-workspace/logs/openclaw.log

15. GPU Optimization (Optional)

Install OpenCL: Intel (intel-opencl-icd) or NVIDIA (nvidia-opencl-dev). Check with clinfo.

16. Docker Deployment

version: '3'
services:
 openclaw:
 image: openclaw/openclaw:latest
 container_name: openclaw
 restart: always
 ports:
 - "8080:8080"
 volumes:
 - ./workspace:/workspace
 environment:
 - OPENAI_API_KEY=YOUR_API_KEY

Run: docker compose up -d

17. Security Hardening

Item	Recommendation
User	Avoid root
Firewall	Enable UFW
SSH	Disable password login
API Keys	Store in .env
Plugins	Install only trusted
Workspace	Sandbox execution
Logs	Audit frequently

18. Common Problems

• Tesseract Not Found: sudo apt install tesseract-ocr
• Missing Language: install tesseract-ocr-ind or tesseract-ocr-mya
• OCR Accuracy Bad: improve DPI, contrast, preprocess image
• OpenClaw Crash: debug with openclaw --debug and check logs

19. Recommended Production Stack

Reverse Proxy: Nginx | SSL: Let's Encrypt | Container: Docker | Process Manager: systemd | OCR: Tesseract | Queue: Redis | Database: PostgreSQL | Monitoring: Prometheus+Grafana

20. Useful Operational Commands

Function	Command
Start	openclaw
Debug	openclaw --debug
Check plugins	openclaw plugins list
OCR test	tesseract img.png stdout
Multi-lang OCR	tesseract img.png stdout -l eng+ind+mya
Logs	journalctl -u openclaw -f

21. Example Complete Workflow

OCR Invoice Automation: Upload Invoice → OCR Extraction → Text Cleaning → AI Classification → JSON Output → Spreadsheet Export → ERP Integration

22. Advanced Integration Ideas

Selenium (browser automation), OpenCV (image preprocessing), Whisper (STT), MongoDB, PostgreSQL, Redis, Telegram Bot, WhatsApp Gateway.

23. Best Practices

• Use Docker for easier deployment
• Separate workspace for isolation
• Use GPU for faster OCR
• Preprocess images for higher accuracy
• Use systemd for auto-restart
• Monitor logs and limit plugin trust

24. Example Full Ubuntu Installation Bundle

sudo apt update && sudo apt upgrade -y
sudo apt install -y python3 python3-pip git curl vim ffmpeg imagemagick poppler-utils docker.io docker-compose tesseract-ocr libtesseract-dev tesseract-ocr-eng tesseract-ocr-ind tesseract-ocr-mya
sudo systemctl enable docker && sudo systemctl start docker

25. Recommended Next Step

Enable OCR pipeline → Integrate Telegram/WhatsApp → Add Selenium automation → Vector memory database → Configure persistent agents → Build workflow automation → Monitoring + backup → Deploy behind Nginx reverse proxy.

---

⚡ Production ready OpenClaw on Ubuntu: optimised, secure and scalable.

1. Verifikasi Instalasi OpenClaw

Pastikan OpenClaw terinstal dengan benar dan dapat diakses.

openclaw --version

atau:

which openclaw

Hasil yang diharapkan: /usr/local/bin/openclaw

2. Lingkungan Ubuntu yang Direkomendasikan

Komponen	Rekomendasi
Ubuntu	22.04 LTS / 24.04 LTS
Python	3.10+
NodeJS	20+
RAM	Minimal 8 GB
CPU	4 core
GPU	Opsional namun disarankan
Docker	Opsional

3. Membuat Direktori Kerja

mkdir -p ~/openclaw-workspace
cd ~/openclaw-workspace

Struktur yang direkomendasikan:

~/openclaw-workspace/
├── projects/
├── uploads/
├── outputs/
├── logs/
├── temp/
├── scripts/
└── models/

4. Konfigurasi OpenClaw

Konfigurasi utama biasanya menggunakan openclaw.json. Buat konfigurasi:

nano openclaw.json

Contoh konfigurasi:

{
 "workspace": "/home/ubuntu/openclaw-workspace",
 "memory": { "provider": "sqlite" },
 "llm": { "provider": "openai", "model": "gpt-4.1" },
 "logging": { "level": "info" },
 "ocr": { "enabled": true, "engine": "tesseract" }
}

5. Mengatur Variabel Lingkungan

nano ~/.bashrc

Tambahkan:

export OPENAI_API_KEY="YOUR_API_KEY"
export OPENCLAW_HOME="$HOME/openclaw-workspace"
export TESSDATA_PREFIX="/usr/share/tesseract-ocr/5/tessdata"

Muat ulang: source ~/.bashrc

6. Instalasi Dependensi OCR (Penting)

sudo apt update
sudo apt install -y tesseract-ocr libtesseract-dev tesseract-ocr-eng tesseract-ocr-ind tesseract-ocr-mya imagemagick poppler-utils ffmpeg

Verifikasi: tesseract --version
Daftar bahasa: tesseract --list-langs
Diharapkan: eng ind mya

7. Konfigurasi OCR Myanmar

sudo apt install tesseract-ocr-mya

Uji OCR: tesseract sample-myanmar.png stdout -l mya
Campuran bahasa: tesseract sample.png stdout -l eng+ind+mya

8. Meningkatkan Akurasi OCR

Pra-pemrosesan yang direkomendasikan:

• Konversi ke skala abu-abu: convert input.png -colorspace Gray output.png
• Tingkatkan kontras: convert input.png -brightness-contrast 20x40 output.png
• Ubah ukuran: convert input.png -resize 300% output.png
• Hilangkan noise: convert input.png -despeckle output.png

9. Mengaktifkan Skill OCR OpenClaw

openclaw plugins install tesseract-ocr

atau openclaw skills install tesseract-ocr
Verifikasi: openclaw plugins list

10. Menjalankan OpenClaw

Mode Interaktif: openclaw
Mode Proyek: openclaw ./projects/myproject
Mode Agen: openclaw agent start
Mode Debug: openclaw --debug

11. Menggunakan OpenClaw dengan OCR

Alur kerja contoh: unggah gambar → ekstraksi OCR → analisis LLM → otomatisasi.
Contoh: openclaw analyze invoice.png
Keluaran: ekstraksi teks, analisis faktur, terjemahan, pembuatan JSON, dll.

12. Integrasi Telegram / WhatsApp

Contoh konfigurasi Telegram:

{ "telegram": { "enabled": true, "bot_token": "TOKEN_ANDA" } }

Jalankan agen: openclaw telegram start
⚠️ Kualitas OCR tergantung kompresi gambar. Disarankan unggah file dan tangkapan layar resolusi tinggi.

13. Menjalankan OpenClaw sebagai Layanan

sudo nano /etc/systemd/system/openclaw.service

Isi:

[Unit]
Description=Layanan OpenClaw
After=network.target

[Service]
Type=simple
User=ubuntu
WorkingDirectory=/home/ubuntu/openclaw-workspace
ExecStart=/usr/local/bin/openclaw
Restart=always

[Install]
WantedBy=multi-user.target

sudo systemctl daemon-reload
sudo systemctl enable openclaw
sudo systemctl start openclaw
sudo systemctl status openclaw

14. Log dan Monitoring

journalctl -u openclaw -f

atau tail -f ~/openclaw-workspace/logs/openclaw.log

15. Optimalisasi GPU (Opsional)

Instal OpenCL: Intel (intel-opencl-icd) atau NVIDIA (nvidia-opencl-dev). Periksa dengan clinfo.

16. Deployment dengan Docker

version: '3'
services:
 openclaw:
 image: openclaw/openclaw:latest
 container_name: openclaw
 restart: always
 ports:
 - "8080:8080"
 volumes:
 - ./workspace:/workspace
 environment:
 - OPENAI_API_KEY=API_KEY_ANDA

Jalankan: docker compose up -d

17. Penguatan Keamanan

Item	Rekomendasi
User	Hindari root
Firewall	Aktifkan UFW
SSH	Nonaktifkan login password
Kunci API	Simpan di .env
Plugin	Instal hanya yang terpercaya
Workspace	Eksekusi dalam sandbox
Log	Audit secara rutin

18. Masalah Umum

• Tesseract tidak ditemukan: sudo apt install tesseract-ocr
• Bahasa hilang: instal tesseract-ocr-ind atau tesseract-ocr-mya
• Akurasi OCR buruk: tingkatkan DPI, kontras, praproses gambar
• OpenClaw crash: debug dengan openclaw --debug dan periksa log

19. Stack Produksi yang Direkomendasikan

Reverse Proxy: Nginx | SSL: Let's Encrypt | Kontainer: Docker | Manajer Proses: systemd | OCR: Tesseract | Antrean: Redis | Database: PostgreSQL | Monitoring: Prometheus+Grafana

20. Perintah Operasional Berguna

Fungsi	Perintah
Mulai	openclaw
Debug	openclaw --debug
Lihat plugin	openclaw plugins list
Uji OCR	tesseract img.png stdout
OCR multi-bahasa	tesseract img.png stdout -l eng+ind+mya
Logs	journalctl -u openclaw -f

21. Contoh Alur Kerja Lengkap

Otomatisasi Faktur OCR: Unggah Faktur → Ekstraksi OCR → Pembersihan Teks → Klasifikasi AI → Output JSON → Ekspor Spreadsheet → Integrasi ERP

22. Ide Integrasi Lanjutan

Selenium (otomatisasi browser), OpenCV (pra-pemrosesan gambar), Whisper (STT), MongoDB, PostgreSQL, Redis, Bot Telegram, Gateway WhatsApp.

23. Praktik Terbaik

• Gunakan Docker untuk deployment lebih mudah
• Pisahkan workspace untuk isolasi
• Gunakan GPU untuk OCR lebih cepat
• Praproses gambar untuk akurasi lebih tinggi
• Gunakan systemd untuk mulai ulang otomatis
• Pantau log dan batasi kepercayaan plugin

24. Paket Instalasi Ubuntu Lengkap (Contoh)

sudo apt update && sudo apt upgrade -y
sudo apt install -y python3 python3-pip git curl vim ffmpeg imagemagick poppler-utils docker.io docker-compose tesseract-ocr libtesseract-dev tesseract-ocr-eng tesseract-ocr-ind tesseract-ocr-mya
sudo systemctl enable docker && sudo systemctl start docker

25. Langkah Selanjutnya yang Direkomendasikan

Aktifkan pipeline OCR → Integrasi Telegram/WhatsApp → Tambahkan otomatisasi Selenium → Database memori vektor → Konfigurasi agen persisten → Bangun alur kerja otomatis → Monitoring + backup → Deployment di belakang reverse proxy Nginx.

---

⚡ Siap produksi OpenClaw di Ubuntu: dioptimalkan, aman, dan skalabel.