Add LinkedIn scraper/chat CLI, Dockerfile, and docker-compose for RAG Gradio UI

Dockerfile

@@ -0,0 +1,34 @@
+FROM python:3.11-slim
+
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    chromium \
+    chromium-driver \
+    fonts-liberation \
+    libasound2 \
+    libatk-bridge2.0-0 \
+    libatk1.0-0 \
+    libcups2 \
+    libdbus-1-3 \
+    libdrm2 \
+    libgbm1 \
+    libgtk-3-0 \
+    libnss3 \
+    libx11-xcb1 \
+    libxcomposite1 \
+    libxdamage1 \
+    libxfixes3 \
+    libxrandr2 \
+    xdg-utils \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+CMD ["python", "linkedin_tool.py", "chat", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,10 +1,93 @@
 # OSINT Tools
-Before running the project, copy .env.default to .env, then update your LinkedIn and OpenAI credentials and API keys in the .env file.
 
-## LinkedIn Posts Analyser
-This is a Jupyter Notebook project. The file linkedin.ipynb is a concept for a LinkedIn user post scraper and GenAI analyzer.
-This tool is useful for OSINT purposes or for managing your digital footprint. It can retrieve LinkedIn posts from several years back and export them to a JSON file.
+This repository now includes a **Python script version** of the original Jupyter notebook (`linkedin.ipynb`).
 
-## Chat with GenAI
-For the GenAI features, an OpenAI API key is required. In this part, the JSON file is imported into ChromaDB. Then, using the Gradio chat interface, an OpenAI chatbot is loaded with knowledge about the gathered LinkedIn posts.
+## 1) Setup environment variables
 
+Copy `.env.default` to `.env` and fill in your credentials/API key:
+
+```bash
+cp .env.default .env
+```
+
+Required variables in `.env`:
+
+- `LINKEDIN_USER`
+- `LINKEDIN_PASSWORD`
+- `LINKEDIN_TARGET_USERNAME`
+- `LINKEDIN_TARGET_NAME`
+- `OPENAI_API_KEY`
+
+> Keep `.env` private. It is already ignored by git.
+
+---
+
+## 2) Run as Python script
+
+Main script: `linkedin_tool.py`
+
+### Scrape LinkedIn posts
+
+```bash
+python linkedin_tool.py scrape
+```
+
+If you need a visible browser window (instead of headless):
+
+```bash
+python linkedin_tool.py scrape --headed
+```
+
+This generates `posts.json`.
+
+### Start RAG chat UI (Gradio)
+
+```bash
+python linkedin_tool.py chat --host 0.0.0.0 --port 7860
+```
+
+### Run scrape + chat in sequence
+
+```bash
+python linkedin_tool.py all
+```
+
+---
+
+## 3) Run with Docker Compose
+
+Build and start the services:
+
+```bash
+docker compose up --build
+```
+
+Docker Compose now starts:
+
+- `osint` (Python app)
+- `selenium` (Selenium + Chrome browser infrastructure)
+
+By default, chat is available at http://localhost:7860 and Selenium Grid at http://localhost:4444.
+
+### Run scraping inside container
+
+```bash
+docker compose run --rm osint python linkedin_tool.py scrape
+```
+
+The `osint` container is preconfigured to use the Selenium browser service via `SELENIUM_REMOTE_URL=http://selenium:4444/wd/hub`.
+
+### Start chat after scraping
+
+```bash
+docker compose up
+```
+
+---
+
+## Notes
+
+- Cookies are stored under `cookies/`.
+- Embedded vector data is stored under `chromadb/`.
+- Scraped posts are saved to `posts.json`.
+- LinkedIn may trigger checkpoint/verification challenges. If that happens, complete verification and rerun.

docker-compose.yml

@@ -0,0 +1,23 @@
+services:
+  osint:
+    build: .
+    container_name: osint-linkedin
+    env_file:
+      - .env
+    environment:
+      - SELENIUM_REMOTE_URL=http://selenium:4444/wd/hub
+    volumes:
+      - ./:/app
+    ports:
+      - "7860:7860"
+    depends_on:
+      - selenium
+    command: ["python", "linkedin_tool.py", "chat", "--host", "0.0.0.0", "--port", "7860"]
+
+  selenium:
+    image: selenium/standalone-chrome:latest
+    container_name: osint-selenium
+    shm_size: 2gb
+    ports:
+      - "4444:4444"
+      - "7900:7900"