OCI-MCP-Servers
OCI MCP Servers
Documentation
OCI MCP Servers
This repository provides a collection of Multi-tool Control Protocol (MCP) servers for managing Oracle Cloud Infrastructure (OCI) resources. Each server exposes a set of tools for a specific OCI service, such as Autonomous Database (ADB), Compute, Identity, and Networking.
These servers are designed to be integrated into development environments like Cursor to allow natural language-based interaction with your OCI tenancy.
Prerequisites
Before you begin, ensure you have the following:
- Python 3.9 or higher.
- An Oracle Cloud Infrastructure account.
- A configured OCI CLI profile. The scripts default to using the `[DEFAULT]` profile located in `~/.oci/config`.
Installation
1. Clone the repository:
git clone https://github.com/shjanjua/OCI-MCP-Servers
cd ocimcp2. Create and activate a virtual environment (recommended):
python -m venv .venv
source .venv/bin/activate
# On Windows, use: .venv\Scripts\activate3. Install the required dependencies:
Each service has its own `requirements.txt`, but they are all the same.
pip install -r compute/requirements.txtUsage in Cursor
You can integrate these tools directly into the Cursor IDE, allowing the AI to use them to perform actions in your OCI tenancy. There are two methods for this: `stdio` and `streamable-http`.
Method 1: Standard I/O (stdio)
This method is the simplest, where Cursor manages the lifecycle of the server process. You configure Cursor to run the `server_stdio.py` script for each service.
Cursor JSON Configuration (`~/.cursor/mcp.json`):
To configure a server, add the following to your `mcpServers` object in Cursor's settings.
{
"mcpServers": {
"adb-stdio": {
"command": "/path/to/your/.venv/bin/python",
"args": ["-u", "/path/to/your/ocimcp/adb/server_stdio.py"]
},
"compute-stdio": {
"command": "/path/to/your/.venv/bin/python",
"args": ["-u", "/path/to/your/ocimcp/compute/server_stdio.py"]
},
"identity-stdio": {
"command": "/path/to/your/.venv/bin/python",
"args": ["-u", "/path/to/your/ocimcp/identity/server_stdio.py"]
},
"network-stdio": {
"command": "/path/to/your/.venv/bin/python",
"args": ["-u", "/path/to/your/ocimcp/network/server_stdio.py"]
}
}
}Note: Replace `/path/to/your/` with the absolute path to your project directory and virtual environment.
Method 2: Streamable HTTP
This method runs each service as a separate, persistent web server. This allows multiple clients to connect simultaneously.
1. Run the Servers:
You must start each server manually in its own terminal window. They are configured to run on different ports to avoid conflicts.
- Terminal 1 (ADB):
python adb/server.py- Terminal 2 (Compute):
python compute/server.py- Terminal 3 (Identity):
python identity/server.pyTerminal 4 (Network):
python network/server.py2. Cursor JSON Configuration (`~/.cursor/mcp.json`):
Add the following configuration to your `mcpServers` object. This tells Cursor where to find the running MCP servers.
{
"mcpServers": {
"adb": {
"url": "http://0.0.0.0:8000/mcp"
},
"compute": {
"url": "http://0.0.0.0:8001/mcp"
},
"identity": {
"url": "http://0.0.0.0:8002/mcp"
},
"network": {
"url": "http://0.0.0.0:8003/mcp"
}
}
}Frequently asked questions
What is OCI-MCP-Servers?
OCI-MCP-Servers is OCI MCP Servers
How do I install OCI-MCP-Servers?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is OCI-MCP-Servers open source?
Yes — it is hosted on GitHub at https://github.com/shjanjua/OCI-MCP-Servers and has 2 stars.
Related MCP tools
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
Automate browser based workflows with AI
Hindsight: Agent Memory That Learns
A privacy-first app that strips AI watermarks from content you own.
Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.
The power of Claude Code / GeminiCLI / CodexCLI + [Gemini / OpenAI / OpenRouter / Azure / Grok / Ollama / Custom Model / All Of The Above] working as one.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP