Disclosure

This is using ChatGPT to generate for almost all of the repository. Including the README.md, except this line.

💻 Ollama SQLite3 CLI Chat: Local AI Chat in The CLI

A lightweight, terminal-based chat interface using Ollama as the AI model backend, sqlite3 for persistent chat storage, and Markdown for readable session logs.

Features

• Chat with local LLMs via Ollama (http://localhost:11434)
• Chats stored in SQLite database and automatically exported to Markdown
• Organized by folder & filename (e.g. chats/default/chat-1.md)
• Supports switching models mid-conversation (/switch)
• Timezone-aware timestamps in messages
• Command-line menu interface

---

Requirements

• Python 3.9+
• Ollama installed and running
• A model pulled via Ollama (e.g. llama3, mistral, etc.)

---

Installation

git clone https://github.com/Corfliss/local-memory-ollama-cli-chat.git
cd local-memory-ollama-cli-chat
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

Example of requirements.txt:

requests
python-dotenv

Configuration

Create a .env file in the root directory to customize the settings:

OLLAMA_MODEL=llama3
OLLAMA_HOST=http://localhost:11434
DB_PATH=database/app.db
TZ_OFFSET=7

• OLLAMA_MODEL: Default model name (e.g., mistral, llama3)
• OLLAMA_HOST: Ollama server URL
• DB_PATH: Path to SQLite DB file
• TZ_OFFSET: Timezone offset from GMT (e.g., 7 for GMT+7)

Running the Program:

Use the CLI menu to interact on /scripts:

python run.py

Menu Options:

1. Start new chat
2. List existing chats
3. Continue a chat
4. Export all chats to markdown
5. Exit

Commands During Chat

• /switch <model> — switch model mid-chat. Example: /switch mistral
• exit or quit — end the chat

Folder Structure

ollama-cli-chat/
├── chats/                  # Markdown logs organized by folder
│   └── default/
│       └── chat-1.md
├── database/               # SQLite DB location (default path)
│   └── app.db
├── chat.py                 # Core chat logic
├── db.py                   # Database logic
├── export.py               # Markdown export utility
├── interface.py            # CLI menu (formerly run.py)
├── requirements.txt
└── README.md

Example Chat Output (Markdown)

Chat: example-chat

Started: 2025.07.14 23:45:02 +0700

---

User — 2025.07.14 23:45:02 +0700

How do I write a recursive function in Python?

Assistant — 2025.07.14 23:45:05 +0700 — llama3

A recursive function is one that calls itself. Here's an example...

Dev Notes

• All messages are stored in the database and reflected in Markdown in real time
• Changing models mid-conversation updates both markdown and database with model info