Individual Research System
A core that can be connected to an LLM, a website, or any other system.
No hallucinations
This is not an LLM model — data is stored in structured cells, and you can see exactly where and how your data exists. Hallucinations can only happen on the LLM side. See structure here (#datastructure).
Not a chatbot inside the system
The system does not use built-in chatbots (Gemini, ChatGPT, etc.), so you don’t pay tokens inside the platform. It is an API visible to ChatGPT or Gemini — you access your data as an external service from your own subscription. Chatbots act as a smart interface to Specus API. Example: “Add an insight and link it to my new project.”
Control through ChatGPT
This is a real CRUD system. You can add information to your database through external access, including using ChatGPT with voice messages. You can capture ideas while walking, and later analyze, structure, and work with them inside your system.
Deploy on a VPS in minutes
You can deploy it on a VPS server in minutes — even the simplest plan is enough. Clone the project, install dependencies, run migrations, and start the server.
git clone https://github.com/specusgear/specus-api.git
cd specus-api
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python manage.py migrate
python manage.py runserverCore system layout
This public release contains only the essential system layer. Domain-specific modules are excluded so the repository stays clean, reusable, and easy to understand.
config/ → Django configuration
core/ → models, serializers, views, schema
manage.py → project entrypoint
docs/ → setup, architecture, api, development, deploymentExplicit model visualization
This is not hidden inside an LLM. The structure is explicit: models, fields, and relations are visible as a real schema.
As I see it: who this product is for and why it’s important to have a basic understanding of such systems
Simple and understandable
This is a basic system that anyone can understand, even without a technical background.
Not a niche tool
This is not a tool only for developers — it reflects a general way of structuring data and systems.
Modern literacy
Understanding systems like this is becoming a minimal standard for a modern person working with information and technology.
About how tomorrow may look and independence from platforms
Switching between LLMs
Today you use Gemini — tomorrow you move to ChatGPT. Nothing is lost. Your knowledge is not tied to a chat history. You keep everything in your own system and simply change the interface.
Loss of access
You are doing research in Europe — and suddenly access to major AI systems is restricted. ChatGPT, Gemini — no longer available. Your knowledge base remains intact. You continue working using alternative or local LLMs, connecting them to the same hub: Specus API.
Ready for the next layer
This version is simple on purpose. It already includes the base structure and can be extended with authentication, API access, SDKs, and other features when needed.