OCI MCP Servers

OCI MCP Servers is an MCP server collection that connects AI assistants to Oracle Cloud Infrastructure (OCI) resources. Designed for cloud engineers, DevOps professionals, and system administrators, this project allows developers to manage cloud environments directly through natural language interfaces in environments such as the Cursor IDE. By breaking functionalities into dedicated modules, the system exposes distinct tools for key OCI service domains including Autonomous Database (ADB), Compute instances, Identity and Access Management (IAM), and Virtual Cloud Network (VCN) components. Users can query the status of running instances, inspect network setups, check tenancy identities, and interact with autonomous database instances without needing to manually run manual OCI CLI commands or switch contexts to the web console. The implementation supports communication through standard input/output (stdio) pipelines managed directly by client environments or persistent HTTP streamable web servers on dedicated ports, offering flexible local and remote configurations backed by the official local OCI CLI configuration profile.

Category: Cloud & Infrastructure

Tags: cloud, cloud-management, infrastructure, oci, oracle cloud

Visit OCI MCP Servers

How to install and configure OCI MCP Servers

1. Clone the repository and navigate into the folder: bash git clone https://github.com/shjanjua/OCI-MCP-Servers cd ocimcp 2. Create and activate a virtual environment: bash python -m venv .venv source .venv/bin/activate 3. Install the required dependencies: bash pip install -r compute/requirements.txt 4. Ensure you have an active OCI CLI profile configured at ~/.oci/config using the [DEFAULT] profile. 5. To configure standard I/O in Cursor (~/.cursor/mcp.json), add the server paths under mcpServers: json { "mcpServers": { "compute-stdio": { "command": "/path/to/your/.venv/bin/python", "args": ["-u", "/path/to/your/ocimcp/compute/server_stdio.py"] } } } Alternatively, run python compute/server.py and register the HTTP endpoint http://0.0.0.0:8001/mcp.

What you can do with OCI MCP Servers

• Querying and managing Oracle Cloud Infrastructure Compute instances directly from your development workspace. - Automating routine operations and status checks across Oracle Autonomous Database (ADB) environments. - Inspecting Virtual Cloud Network (VCN) components, subnets, and routing configurations via natural language queries. - Auditing Identity and Access Management (IAM) configurations, users, and groups within your OCI tenancy.

Key facts

• https://github.com/shjanjua/OCI-MCP-Servers
• Cloud & Infrastructure
• cloud, cloud-management, infrastructure, oci, oracle cloud

Part of MCP Servers

What is OCI MCP Servers?

OCI MCP Servers is a collection of Model Context Protocol servers designed to bridge Oracle Cloud Infrastructure services with AI developer tools like Cursor. It exposes dedicated tools for OCI Autonomous Database, Compute, Identity, and Networking modules.

What are the prerequisites for running OCI MCP Servers?

You need Python 3.9 or higher, an active Oracle Cloud Infrastructure account, and a configured OCI CLI configuration file located at ~/.oci/config with a valid DEFAULT profile.

Which MCP clients work with OCI MCP Servers?

OCI MCP Servers is primarily documented for use with the Cursor IDE. It supports standard I/O (stdio) process execution as well as streamable HTTP server endpoints, making it compatible with any MCP client supporting these connection types.

How do the HTTP and stdio modes differ in OCI MCP Servers?

Standard I/O mode lets the MCP client spawn and control the Python server process automatically. The streamable HTTP mode runs persistent background web servers across separate local ports, enabling multiple clients to connect at the same time.