Skip to content

System-Requirement-Checks.md

Latest commit

 

History

History
111 lines (53 loc) · 2.93 KB

System-Requirement-Checks.md

File metadata and controls

111 lines (53 loc) · 2.93 KB

🛠️ System Requirements & Dependency Verification

Before running the Sovereign AI Agent, use this guide to ensure all required software, AI engines, and dependencies are correctly installed on your Ubuntu machine.

  1. The AI Engine (Ollama)

Ollama is the core service that runs the Large Language Models locally on your CPU/GPU.

Installation Command:

curl -fsSL https://ollama.com/install.sh | sh

Verification Check 1 (Software):

ollama -v

Expected Outcome: You should see something like ollama version is 0.1.x. If you see this, the core engine is installed.

Verification Check 2 (AI Models): Ensure the specific models are downloaded:

ollama list

Expected Outcome: The table output should list llama3.2:3b and qwen2.5:3b.

Fix if missing: Run ollama pull llama3.2:3b and ollama pull qwen2.5:3b.

Verification Check 3 (Background Service):

curl http://localhost:11434

Expected Outcome: It should reply with Ollama is running. If you get a "Connection refused" error, you need to start it by typing ollama serve.

  1. Frontend Environment (Node.js & npm)

Required to run the Vite/React dashboard. As we discovered earlier, Vite requires Node version 20 or higher.

Verification Check:

node -v npm -v

Expected Outcome: node -v must return v20.x.x or higher. npm -v should return 9.x.x or higher.

Fix if missing/outdated:

curl -fsSL https://deb.nodesource.com/setup_20.x | sudo -E bash - sudo apt-get install -y nodejs

  1. Backend Environment (Python 3)

Required for the FastAPI server, JobSpy scraping, and Browser-Use automation.

Verification Check 1 (Python Engine):

python3 --version

Expected Outcome: Python 3.10.x or higher (Ubuntu 22.04 usually comes with 3.10).

Verification Check 2 (Virtual Environment tool):

dpkg -l | grep python3-venv

Expected Outcome: You should see ii python3-venv in the list.

Fix if missing: Run sudo apt install python3-venv python3-pip.

  1. Browser Automation (Playwright)

Playwright is the physical browser engine that browser-use controls to click "Apply" on websites.

Verification Check: First, ensure you are inside your activated Python backend virtual environment (source venv/bin/activate), then run:

playwright --version

Expected Outcome: Version 1.4x.x

Fix if missing/error: You need to install the browser binaries. Run:

pip install playwright playwright install

(Note: This downloads about 200MB of Chromium browser data so the AI has a clean sandbox to browse the web).

  1. Port Conflict Safety Check

The agent requires three specific network ports to communicate internally on your machine. Ensure nothing else is blocking them.

Verification Check:

sudo lsof -i -P -n | grep LISTEN

Expected Outcome: * Port 11434 should be used by ollama.

Port 8000 should be free (until you start agent_backend.py).

Port 5173 should be free (until you start npm run dev).

If 8000 or 5173 are in use by other projects, stop those projects first.