What is Geminimcp

GeminiMCP is a command-line toolset designed for interacting with Google Gemini LLM (Large Language Model) APIs, enabling users to send prompts, generate Linux commands from natural language, and manage logs in a local SQLite database.

Use cases

Use cases for GeminiMCP include automating command-line tasks, simplifying interactions with APIs, and logging command execution results for auditing and analysis.

How to use

To use GeminiMCP, clone the repository, set up a virtual environment, install dependencies, obtain a Google Gemini API key, configure the database and model settings, and then initialize the database. You can send prompts and execute commands through the command line.

Key features

Key features include sending prompts to Google Gemini LLM, generating and executing Linux commands from natural language, logging sessions and command results in a local SQLite database, and managing prompt templates via CLI tools.

Where to use

GeminiMCP can be used in various fields such as software development, system administration, and automation tasks where natural language processing and command execution are beneficial.

GeminiMCP

Overview

GeminiMCP is a command-line toolset for interacting with Google Gemini LLM (Large Language Model) APIs.
It allows you to send prompts, generate and execute Linux commands from natural language, and manage prompt/response logs in a local SQLite database.
The project is modular and extensible, making it easy to customize for your own workflow.

Features

• Send prompts to Google Gemini LLM and receive responses
• Generate Linux commands from natural language instructions and execute them
• Log sessions, prompts, responses, and command results to a local SQLite database
• Manage prompt templates and database contents via CLI tools

Initial Setup

1. Clone the repository

git clone https://github.com/yourname/GeminiMCP.git
cd GeminiMCP

2. Create and activate a virtual environment

python3 -m venv venv
source venv/bin/activate   # On Windows: venv\Scripts\activate

3. Install dependencies

pip install -r requests.txt

4. Obtain a Google Gemini API Key

• Visit: https://ai.google.dev/gemini-api/docs/api-key?hl=ja
• Copy your API key.

5. Create a .env file in the project root

API_KEY=your-api-key-here

6. Configure the database and model

Edit config/config.yaml as needed (default settings should work for most cases):

model_name: "gemini-2.5-flash-preview-05-20"
db_file: "./data/db.db"
set_up: "setup.sql"

7. Initialize the database

python src/init_db.py

Usage

Send a prompt to Gemini

python src/send_prompt.py

Run MCP (natural language → command execution)

python src/mcp_execute.py

Manage the database (view sessions, prompts, etc.)

python src/manage_db.py --help

Notes

• .env and all .db files are excluded from git via .gitignore.
• For development, always use a virtual environment (venv).