🔥 OpenClaw OCR Automation MVP

2 Docker Containers · FastAPI + Tesseract + Watchdog

🚁 Helicopter View — Whole System at a Glance

MVP Architecture

What are we building? A lightweight OCR automation pipeline using two independent Docker containers.

📦 Container 1 (OCR API): FastAPI web server with Tesseract OCR engine. Exposes /ocr endpoint to accept image uploads and returns extracted text as JSON.
📦 Container 2 (Watcher Agent): A Python watchdog script that monitors a local folder (./uploads). Whenever a new image appears, it automatically sends a POST request to the OCR API and prints the result.

🧩 Data flow: User copies image → uploads/ → Watcher detects → HTTP POST to OCR API → Tesseract processes → JSON text output.
✅ Why this MVP? Fully containerized, portable, scalable, and can evolve into an intelligent document processing system with AI agents later.

+------------------+ +------------------+ +------------------+ +------------------+ | Upload Folder | ---> | Watcher Agent | ---> | OCR API | ---> | JSON Result | | ./uploads | | (watchdog) | | FastAPI+Tesseract| | Extracted Text | +------------------+ +------------------+ +------------------+ +------------------+

📘 Step-by-Step Implementation & Code Explanations

28 Steps Detailed

1-4 🐳 Docker Installation & Setup

Install Docker Engine and Compose, enable service, add user to docker group, verify versions.

sudo apt update && sudo apt install -y docker.io docker-compose-plugin
sudo systemctl enable docker && sudo systemctl start docker
sudo usermod -aG docker $USER && newgrp docker
docker --version && docker compose version

5-7 📁 Project Structure

Create root folders and subfolders for OCR API and watcher agent.

cd ~/.openclaw && mkdir -p openclaw-ocr-mvp openclaw-agent
cd openclaw-ocr-mvp && mkdir uploads outputs

8 📄 requirements.txt (OCR API)

List of Python dependencies: FastAPI, Uvicorn, Pillow for images, pytesseract bridge, and NumPy.

fastapi
uvicorn[standard]
python-multipart
pillow
pytesseract
numpy

9 🖥️ main.py (OCR Service)

FastAPI app with /ocr endpoint. Receives image, saves temporarily, runs OCR via Tesseract, returns JSON. Root endpoint for health check.

from fastapi import FastAPI, UploadFile, File
from PIL import Image
import pytesseract, tempfile, shutil, os

app = FastAPI()

@app.get("/")
async def root():
 return {"status": "running"}

@app.post("/ocr")
async def do_ocr(file: UploadFile = File(...)):
 ext = os.path.splitext(file.filename)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return {"error": "Only image files supported"}
 with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as tmp:
 shutil.copyfileobj(file.file, tmp)
 path = tmp.name
 image = Image.open(path)
 text = pytesseract.image_to_string(image)
 return {"filename": file.filename, "text": text}

10 🐳 Dockerfile (OCR API)

Build image from Python 3.11, install Tesseract engine, copy requirements, install pip dependencies, expose port 8000, run Uvicorn.

FROM python:3.11
WORKDIR /app
RUN apt-get update && apt-get install -y tesseract-ocr
COPY requirements.txt .
RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 8000
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

11 🧩 docker-compose.yml (OCR API)

Simplifies orchestration: builds container, maps port 8000, mounts ./uploads and ./outputs volumes for persistence.

services:
 ocr-api:
 build: .
 container_name: ocr-api
 ports:
 - "8000:8000"
 volumes:
 - ./uploads:/app/uploads
 - ./outputs:/app/outputs
 restart: unless-stopped

12-14 🔨 Build & Run OCR API

cd ~/.openclaw/openclaw-ocr-mvp
docker compose build --no-cache
docker compose up -d
curl http://localhost:8000 # Expected: {"status":"running"}

15-17 👁️ Watcher Agent (requirements + watcher.py)

requirements.txt: watchdog (filesystem events), requests (HTTP calls). watcher.py: monitors /data/inbox, sends new images to OCR API.

# requirements.txt (inside openclaw-agent)
watchdog
requests

# watcher.py
from watchdog.observers import Observer
from watchdog.events import FileSystemEventHandler
import requests, time, os

WATCH_FOLDER = "/data/inbox"
OCR_API = "http://host.docker.internal:8000/ocr"

class Handler(FileSystemEventHandler):
 def on_created(self, event):
 if event.is_directory:
 return
 path = event.src_path
 ext = os.path.splitext(path)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return
 print(f"[NEW FILE] {path}")
 with open(path, "rb") as f:
 files = {"file": f}
 response = requests.post(OCR_API, files=files)
 print(response.json())

if __name__ == "__main__":
 os.makedirs("/data/inbox", exist_ok=True)
 observer = Observer()
 observer.schedule(Handler(), "/data/inbox", recursive=False)
 observer.start()
 print("Watching folder...")
 try:
 while True:
 time.sleep(1)
 except KeyboardInterrupt:
 observer.stop()
 observer.join()

18-19 🐳 Dockerfile & Build Watcher

Dockerfile for watcher: Python 3.11, copy requirements, install, run watcher.py. Build as openclaw-agent.

FROM python:3.11
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
CMD ["python", "watcher.py"]

cd ~/.openclaw/openclaw-agent
docker build -t openclaw-agent .

20 🚀 Run Watcher Container (with mount)

Bind mount the uploads folder from OCR API project into the container at /data/inbox. Use --add-host so container can reach host's OCR API.

docker run -it --name openclaw-agent -v ~/.openclaw/openclaw-ocr-mvp/uploads:/data/inbox --add-host=host.docker.internal:host-gateway openclaw-agent

You should see: Watching folder...

21-22 🧪 Test Automation Workflow

Copy any image into the uploads folder. The watcher detects and prints OCR result automatically.

cp ~/Downloads/sample.jpg ~/.openclaw/openclaw-ocr-mvp/uploads/

Expected watcher output: [NEW FILE] /data/inbox/sample.jpg followed by JSON with extracted text.

23-27 🔄 Container Management

docker ps # see both containers running
docker compose restart # inside ocr-api folder
docker restart openclaw-agent
docker stop openclaw-agent # stop watcher

28 🎯 Final Architecture Status

Both containers run independently. OCR API listens on port 8000, watcher monitors uploads folder. This MVP is ready to be extended with AI reasoning, document classification, or multi-agent orchestration.

🚁 Gambaran Helikopter — Sistem Secara Menyeluruh

Arsitektur MVP

Apa yang kita bangun? Pipeline OCR otomatis ringan menggunakan dua container Docker independen.

📦 Container 1 (OCR API): Server web FastAPI dengan mesin Tesseract OCR. Menyediakan endpoint /ocr untuk menerima upload gambar dan mengembalikan teks hasil ekstraksi dalam format JSON.
📦 Container 2 (Watcher Agent): Script Python watchdog yang memantau folder lokal (./uploads). Setiap ada gambar baru, secara otomatis mengirimkan permintaan POST ke OCR API dan mencetak hasilnya.

🧩 Alur data: Pengguna menyalin gambar → folder uploads/ → Watcher mendeteksi → HTTP POST ke OCR API → Tesseract memproses → output teks JSON.
✅ Mengapa MVP ini? Sepenuhnya containerized, portabel, skalabel, dan bisa berkembang menjadi sistem pemrosesan dokumen cerdas dengan agen AI nantinya.

+------------------+ +------------------+ +------------------+ +------------------+ | Folder Upload | ---> | Agen Watcher | ---> | OCR API | ---> | Hasil JSON | | ./uploads | | (watchdog) | | FastAPI+Tesseract| | Teks Terekstrak | +------------------+ +------------------+ +------------------+ +------------------+

📘 Langkah demi Langkah & Penjelasan Kode

28 Langkah Terperinci

1-4 🐳 Instalasi Docker & Setup

Instal Docker Engine dan Compose, aktifkan layanan, tambahkan user ke grup docker, verifikasi versi.

5-7 📁 Struktur Proyek

Buat folder utama dan subfolder untuk OCR API dan agen watcher.

cd ~/.openclaw && mkdir -p openclaw-ocr-mvp openclaw-agent
cd openclaw-ocr-mvp && mkdir uploads outputs

8 📄 requirements.txt (OCR API)

Daftar dependensi Python: FastAPI, Uvicorn, Pillow untuk gambar, pytesseract sebagai jembatan OCR, dan NumPy.

fastapi
uvicorn[standard]
python-multipart
pillow
pytesseract
numpy

9 🖥️ main.py (Layanan OCR)

Aplikasi FastAPI dengan endpoint /ocr. Menerima gambar, menyimpan sementara, menjalankan OCR via Tesseract, mengembalikan JSON. Endpoint root untuk pengecekan kesehatan.

from fastapi import FastAPI, UploadFile, File
from PIL import Image
import pytesseract, tempfile, shutil, os

app = FastAPI()

@app.get("/")
async def root():
 return {"status": "running"}

@app.post("/ocr")
async def do_ocr(file: UploadFile = File(...)):
 ext = os.path.splitext(file.filename)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return {"error": "Hanya file gambar yang didukung"}
 with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as tmp:
 shutil.copyfileobj(file.file, tmp)
 path = tmp.name
 image = Image.open(path)
 text = pytesseract.image_to_string(image)
 return {"filename": file.filename, "text": text}

10 🐳 Dockerfile (OCR API)

Membangun image dari Python 3.11, menginstal mesin Tesseract, menyalin requirements, menginstal dependensi pip, mengekspos port 8000, menjalankan Uvicorn.

FROM python:3.11
WORKDIR /app
RUN apt-get update && apt-get install -y tesseract-ocr
COPY requirements.txt .
RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 8000
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

11 🧩 docker-compose.yml (OCR API)

Memudahkan orkestrasi: build container, memetakan port 8000, mount volume ./uploads dan ./outputs untuk persistensi.

services:
 ocr-api:
 build: .
 container_name: ocr-api
 ports:
 - "8000:8000"
 volumes:
 - ./uploads:/app/uploads
 - ./outputs:/app/outputs
 restart: unless-stopped

12-14 🔨 Build & Jalankan OCR API

cd ~/.openclaw/openclaw-ocr-mvp
docker compose build --no-cache
docker compose up -d
curl http://localhost:8000 # Diharapkan: {"status":"running"}

15-17 👁️ Agen Watcher (requirements + watcher.py)

requirements.txt: watchdog (event filesystem), requests (panggilan HTTP). watcher.py: memantau /data/inbox, mengirim gambar baru ke OCR API.

# requirements.txt (di dalam folder openclaw-agent)
watchdog
requests

# watcher.py
from watchdog.observers import Observer
from watchdog.events import FileSystemEventHandler
import requests, time, os

WATCH_FOLDER = "/data/inbox"
OCR_API = "http://host.docker.internal:8000/ocr"

class Handler(FileSystemEventHandler):
 def on_created(self, event):
 if event.is_directory:
 return
 path = event.src_path
 ext = os.path.splitext(path)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return
 print(f"[FILE BARU] {path}")
 with open(path, "rb") as f:
 files = {"file": f}
 response = requests.post(OCR_API, files=files)
 print(response.json())

if __name__ == "__main__":
 os.makedirs("/data/inbox", exist_ok=True)
 observer = Observer()
 observer.schedule(Handler(), "/data/inbox", recursive=False)
 observer.start()
 print("Memantau folder...")
 try:
 while True:
 time.sleep(1)
 except KeyboardInterrupt:
 observer.stop()
 observer.join()

18-19 🐳 Dockerfile & Build Watcher

Dockerfile untuk watcher: Python 3.11, salin requirements, instal, jalankan watcher.py. Build sebagai openclaw-agent.

FROM python:3.11
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
CMD ["python", "watcher.py"]

cd ~/.openclaw/openclaw-agent
docker build -t openclaw-agent .

20 🚀 Jalankan Container Watcher (dengan mount)

Mount folder uploads dari proyek OCR API ke dalam container di /data/inbox. Gunakan --add-host agar container dapat mengakses OCR API di host.

docker run -it --name openclaw-agent -v ~/.openclaw/openclaw-ocr-mvp/uploads:/data/inbox --add-host=host.docker.internal:host-gateway openclaw-agent

Seharusnya muncul: Memantau folder...

21-22 🧪 Uji Alur Otomatisasi

Salin gambar ke folder uploads. Watcher akan mendeteksi dan mencetak hasil OCR secara otomatis.

cp ~/Downloads/sample.jpg ~/.openclaw/openclaw-ocr-mvp/uploads/

Output watcher yang diharapkan: [FILE BARU] /data/inbox/sample.jpg diikuti oleh JSON dengan teks yang terekstrak.

23-27 🔄 Manajemen Container

docker ps # lihat kedua container berjalan
docker compose restart # di dalam folder ocr-api
docker restart openclaw-agent
docker stop openclaw-agent # hentikan watcher

28 🎯 Status Arsitektur Final

Kedua container berjalan independen. OCR API mendengarkan di port 8000, watcher memantau folder uploads. MVP ini siap diperluas dengan penalaran AI, klasifikasi dokumen, atau orkestrasi multi-agen.

🔥 OpenClaw OCR Automation MVP

2 Docker Containers · FastAPI + Tesseract + Watchdog

🚁 Helicopter View — Whole System at a Glance

MVP Architecture

📘 Step-by-Step Implementation & Code Explanations

28 Steps Detailed

1-4 🐳 Docker Installation & Setup

Install Docker Engine and Compose, enable service, add user to docker group, verify versions.

5-7 📁 Project Structure

Create root folders and subfolders for OCR API and watcher agent.

cd ~/.openclaw && mkdir -p openclaw-ocr-mvp openclaw-agent
cd openclaw-ocr-mvp && mkdir uploads outputs

8 📄 requirements.txt (OCR API)

List of Python dependencies: FastAPI, Uvicorn, Pillow for images, pytesseract bridge, and NumPy.

fastapi
uvicorn[standard]
python-multipart
pillow
pytesseract
numpy

9 🖥️ main.py (OCR Service)

FastAPI app with /ocr endpoint. Receives image, saves temporarily, runs OCR via Tesseract, returns JSON. Root endpoint for health check.

from fastapi import FastAPI, UploadFile, File
from PIL import Image
import pytesseract, tempfile, shutil, os

app = FastAPI()

@app.get("/")
async def root():
 return {"status": "running"}

@app.post("/ocr")
async def do_ocr(file: UploadFile = File(...)):
 ext = os.path.splitext(file.filename)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return {"error": "Only image files supported"}
 with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as tmp:
 shutil.copyfileobj(file.file, tmp)
 path = tmp.name
 image = Image.open(path)
 text = pytesseract.image_to_string(image)
 return {"filename": file.filename, "text": text}

10 🐳 Dockerfile (OCR API)

Build image from Python 3.11, install Tesseract engine, copy requirements, install pip dependencies, expose port 8000, run Uvicorn.

FROM python:3.11
WORKDIR /app
RUN apt-get update && apt-get install -y tesseract-ocr
COPY requirements.txt .
RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 8000
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

11 🧩 docker-compose.yml (OCR API)

Simplifies orchestration: builds container, maps port 8000, mounts ./uploads and ./outputs volumes for persistence.

services:
 ocr-api:
 build: .
 container_name: ocr-api
 ports:
 - "8000:8000"
 volumes:
 - ./uploads:/app/uploads
 - ./outputs:/app/outputs
 restart: unless-stopped

12-14 🔨 Build & Run OCR API

cd ~/.openclaw/openclaw-ocr-mvp
docker compose build --no-cache
docker compose up -d
curl http://localhost:8000 # Expected: {"status":"running"}

15-17 👁️ Watcher Agent (requirements + watcher.py)

requirements.txt: watchdog (filesystem events), requests (HTTP calls). watcher.py: monitors /data/inbox, sends new images to OCR API.

# requirements.txt (inside openclaw-agent)
watchdog
requests

# watcher.py
from watchdog.observers import Observer
from watchdog.events import FileSystemEventHandler
import requests, time, os

WATCH_FOLDER = "/data/inbox"
OCR_API = "http://host.docker.internal:8000/ocr"

class Handler(FileSystemEventHandler):
 def on_created(self, event):
 if event.is_directory:
 return
 path = event.src_path
 ext = os.path.splitext(path)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return
 print(f"[NEW FILE] {path}")
 with open(path, "rb") as f:
 files = {"file": f}
 response = requests.post(OCR_API, files=files)
 print(response.json())

if __name__ == "__main__":
 os.makedirs("/data/inbox", exist_ok=True)
 observer = Observer()
 observer.schedule(Handler(), "/data/inbox", recursive=False)
 observer.start()
 print("Watching folder...")
 try:
 while True:
 time.sleep(1)
 except KeyboardInterrupt:
 observer.stop()
 observer.join()

18-19 🐳 Dockerfile & Build Watcher

Dockerfile for watcher: Python 3.11, copy requirements, install, run watcher.py. Build as openclaw-agent.

FROM python:3.11
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
CMD ["python", "watcher.py"]

cd ~/.openclaw/openclaw-agent
docker build -t openclaw-agent .

20 🚀 Run Watcher Container (with mount)

Bind mount the uploads folder from OCR API project into the container at /data/inbox. Use --add-host so container can reach host's OCR API.

docker run -it --name openclaw-agent -v ~/.openclaw/openclaw-ocr-mvp/uploads:/data/inbox --add-host=host.docker.internal:host-gateway openclaw-agent

You should see: Watching folder...

21-22 🧪 Test Automation Workflow

Copy any image into the uploads folder. The watcher detects and prints OCR result automatically.

cp ~/Downloads/sample.jpg ~/.openclaw/openclaw-ocr-mvp/uploads/

Expected watcher output: [NEW FILE] /data/inbox/sample.jpg followed by JSON with extracted text.

23-27 🔄 Container Management

docker ps # see both containers running
docker compose restart # inside ocr-api folder
docker restart openclaw-agent
docker stop openclaw-agent # stop watcher

28 🎯 Final Architecture Status

🚁 Gambaran Helikopter — Sistem Secara Menyeluruh

Arsitektur MVP

📘 Langkah demi Langkah & Penjelasan Kode

28 Langkah Terperinci

1-4 🐳 Instalasi Docker & Setup

Instal Docker Engine dan Compose, aktifkan layanan, tambahkan user ke grup docker, verifikasi versi.

5-7 📁 Struktur Proyek

Buat folder utama dan subfolder untuk OCR API dan agen watcher.

cd ~/.openclaw && mkdir -p openclaw-ocr-mvp openclaw-agent
cd openclaw-ocr-mvp && mkdir uploads outputs

8 📄 requirements.txt (OCR API)

Daftar dependensi Python: FastAPI, Uvicorn, Pillow untuk gambar, pytesseract sebagai jembatan OCR, dan NumPy.

fastapi
uvicorn[standard]
python-multipart
pillow
pytesseract
numpy

9 🖥️ main.py (Layanan OCR)

Aplikasi FastAPI dengan endpoint /ocr. Menerima gambar, menyimpan sementara, menjalankan OCR via Tesseract, mengembalikan JSON. Endpoint root untuk pengecekan kesehatan.

from fastapi import FastAPI, UploadFile, File
from PIL import Image
import pytesseract, tempfile, shutil, os

app = FastAPI()

@app.get("/")
async def root():
 return {"status": "running"}

@app.post("/ocr")
async def do_ocr(file: UploadFile = File(...)):
 ext = os.path.splitext(file.filename)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return {"error": "Hanya file gambar yang didukung"}
 with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as tmp:
 shutil.copyfileobj(file.file, tmp)
 path = tmp.name
 image = Image.open(path)
 text = pytesseract.image_to_string(image)
 return {"filename": file.filename, "text": text}

10 🐳 Dockerfile (OCR API)

Membangun image dari Python 3.11, menginstal mesin Tesseract, menyalin requirements, menginstal dependensi pip, mengekspos port 8000, menjalankan Uvicorn.

FROM python:3.11
WORKDIR /app
RUN apt-get update && apt-get install -y tesseract-ocr
COPY requirements.txt .
RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 8000
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

11 🧩 docker-compose.yml (OCR API)

Memudahkan orkestrasi: build container, memetakan port 8000, mount volume ./uploads dan ./outputs untuk persistensi.

services:
 ocr-api:
 build: .
 container_name: ocr-api
 ports:
 - "8000:8000"
 volumes:
 - ./uploads:/app/uploads
 - ./outputs:/app/outputs
 restart: unless-stopped

12-14 🔨 Build & Jalankan OCR API

cd ~/.openclaw/openclaw-ocr-mvp
docker compose build --no-cache
docker compose up -d
curl http://localhost:8000 # Diharapkan: {"status":"running"}

15-17 👁️ Agen Watcher (requirements + watcher.py)

requirements.txt: watchdog (event filesystem), requests (panggilan HTTP). watcher.py: memantau /data/inbox, mengirim gambar baru ke OCR API.

# requirements.txt (di dalam folder openclaw-agent)
watchdog
requests

# watcher.py
from watchdog.observers import Observer
from watchdog.events import FileSystemEventHandler
import requests, time, os

WATCH_FOLDER = "/data/inbox"
OCR_API = "http://host.docker.internal:8000/ocr"

class Handler(FileSystemEventHandler):
 def on_created(self, event):
 if event.is_directory:
 return
 path = event.src_path
 ext = os.path.splitext(path)[1].lower()
 if ext not in [".jpg", ".jpeg", ".png"]:
 return
 print(f"[FILE BARU] {path}")
 with open(path, "rb") as f:
 files = {"file": f}
 response = requests.post(OCR_API, files=files)
 print(response.json())

if __name__ == "__main__":
 os.makedirs("/data/inbox", exist_ok=True)
 observer = Observer()
 observer.schedule(Handler(), "/data/inbox", recursive=False)
 observer.start()
 print("Memantau folder...")
 try:
 while True:
 time.sleep(1)
 except KeyboardInterrupt:
 observer.stop()
 observer.join()

18-19 🐳 Dockerfile & Build Watcher

Dockerfile untuk watcher: Python 3.11, salin requirements, instal, jalankan watcher.py. Build sebagai openclaw-agent.

FROM python:3.11
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
CMD ["python", "watcher.py"]

cd ~/.openclaw/openclaw-agent
docker build -t openclaw-agent .

20 🚀 Jalankan Container Watcher (dengan mount)

Mount folder uploads dari proyek OCR API ke dalam container di /data/inbox. Gunakan --add-host agar container dapat mengakses OCR API di host.

docker run -it --name openclaw-agent -v ~/.openclaw/openclaw-ocr-mvp/uploads:/data/inbox --add-host=host.docker.internal:host-gateway openclaw-agent

Seharusnya muncul: Memantau folder...

21-22 🧪 Uji Alur Otomatisasi

Salin gambar ke folder uploads. Watcher akan mendeteksi dan mencetak hasil OCR secara otomatis.

cp ~/Downloads/sample.jpg ~/.openclaw/openclaw-ocr-mvp/uploads/

Output watcher yang diharapkan: [FILE BARU] /data/inbox/sample.jpg diikuti oleh JSON dengan teks yang terekstrak.

23-27 🔄 Manajemen Container

docker ps # lihat kedua container berjalan
docker compose restart # di dalam folder ocr-api
docker restart openclaw-agent
docker stop openclaw-agent # hentikan watcher

28 🎯 Status Arsitektur Final

Kedua container berjalan independen. OCR API mendengarkan di port 8000, watcher memantau folder uploads. MVP ini siap diperluas dengan penalaran AI, klasifikasi dokumen, atau orkestrasi multi-agen.

OpenClaw OCR API & Agent

🔥 OpenClaw OCR Automation MVP

🚁 Helicopter View — Whole System at a Glance

📘 Step-by-Step Implementation & Code Explanations

🚁 Gambaran Helikopter — Sistem Secara Menyeluruh

📘 Langkah demi Langkah & Penjelasan Kode

🔥 OpenClaw OCR Automation MVP

🚁 Helicopter View — Whole System at a Glance

📘 Step-by-Step Implementation & Code Explanations

🚁 Gambaran Helikopter — Sistem Secara Menyeluruh

📘 Langkah demi Langkah & Penjelasan Kode

Comments