A locally-run agentic AI core with direct system access. Runs terminal commands, reads and writes files, and searches the web autonomously. Built on fine-tuned Qwen2.5-0.5B. Use with caution.

ollama run tcoxav/aegis

curl http://localhost:11434/api/chat \
  -d '{
    "model": "tcoxav/aegis",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='tcoxav/aegis',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'tcoxav/aegis',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 3 weeks ago

3 weeks ago

00c07c120ad5 · 986MB ·

model

archqwen2

parameters1.54B

quantizationQ4_K_M

986MB

system

You are AEGIS, an autonomous system automation core created by TCOxAV. You have access to tools. Whe

629B

params

{ "num_ctx": 2048, "num_gpu": 99, "repeat_penalty": 1.1, "temperature": 0.2 }

69B

Readme

AEGIS

An autonomous system automation core. Runs terminal commands, reads and writes files, and searches the web on your behalf.

Use with caution — AEGIS has direct access to your system.

Quick Start

Pull the model:

ollama pull tcoxav/aegis

Then clone the agent files and run:

git clone https://github.com/TCOxAV/AEGIS.git
cd AEGIS
pip install flask requests duckduckgo-search beautifulsoup4

Windows — double click start.bat Linux — run python3 web_ui.py

Open your browser at http://localhost:5000

What it can do

Run terminal commands on your system
Read and write files anywhere on your machine
Search the web via DuckDuckGo
Fetch and read webpage content
Chain multiple actions automatically to complete complex tasks

Requirements
Ollama
Python 3.10+
1.3GB VRAM minimum
4GB RAM minimum

Built by TCOxAV