OCI Registry MCP Server
Expose OCI registry data to LLM-powered applications via the Model Context Protocol.
Key Features
Use Cases
README
OCI Registry MCP Server
An MCP (Model Context Protocol) server that provides tools for querying OCI
registries and image references.
Overview
This project implements an SSE-based MCP server that allows LLM-powered applications to interact with OCI registries. It provides tools for retrieving information about container images, listing tags, and more.
Features
- Get information about OCI images
- List tags for repositories
- Get image manifests
- Get image configs
MCP Tools
The server provides the following MCP tools:
get_image_info
Get information about an OCI image.
Input:
image_ref: The image reference (e.g., docker.io/library/alpine:latest)
Output:
- Image information including digest, size, architecture, OS, creation date, and number of layers
list_tags
List tags for a repository.
Input:
repository: The repository name (e.g., docker.io/library/alpine)
Output:
- List of tags for the repository
get_image_manifest
Get the manifest for an OCI image.
Input:
image_ref: The image reference (e.g., docker.io/library/alpine:latest)
Output:
- The image manifest
get_image_config
Get the config for an OCI image.
Input:
image_ref: The image reference (e.g., docker.io/library/alpine:latest)
Output:
- The image config
Usage
Running with ToolHive (Recommended)
The easiest way to run the OCI Registry MCP server is using ToolHive, which provides secure, containerized deployment of MCP servers:
# Install ToolHive (if not already installed)
# See: https://docs.stacklok.com/toolhive/guides-cli/install
# Register a supported client so ToolHive can auto-configure your environment
thv client setup
# Run the OCI Registry MCP server (packaged as 'oci-registry' in ToolHive)
thv run oci-registry
# List running servers
thv list
# Get detailed information about the server
thv registry info oci-registry
The server will be available to your MCP-compatible clients and can query OCI registries for image information.
Authentication with ToolHive
If you need to access private registries, you can provide authentication credentials using ToolHive's secret management:
# For bearer token authentication
thv secret set oci-token
# Enter your bearer token when prompted
thv run --secret oci-token,target=OCI_TOKEN oci-registry
# For username/password authentication
thv secret set oci-username
thv secret set oci-password
# Enter your credentials when prompted
thv run --secret oci-username,target=OCI_USERNAME --secret oci-password,target=OCI_PASSWORD oci-registry
Development
Prerequisites
- Go 1.21 or later
- Access to OCI registries
Authentication
The server supports the following authentication methods for accessing private OCI registries (in order of priority):
-
HTTP Authorization Header (Highest Priority): Include a bearer token in the HTTP request's
Authorizationheader:Authorization: Bearer <your-token>- This method takes precedence over all other authentication methods
- When present, environment variables and Docker config are ignored
-
Bearer Token Environment Variable: Set the following environment variable:
OCI_TOKEN: Bearer token for registry authentication
-
Username and Password: Set the following environment variables:
OCI_USERNAME: Username for registry authenticationOCI_PASSWORD: Password for registry authentication
-
Docker Config (Lowest Priority): If no other authentication is provided, the server will use the default Docker keychain, which reads credentials from
~/.docker/config.json.
Examples:
# HTTP Authorization header (for per-request authentication)
# This is handled automatically by the MCP client when making requests
# Example: curl -H "Authorization: Bearer mytoken" http://localhost:8080/...
# Bearer token authentication via environment variable
export OCI_TOKEN=mytoken
# Username/password authentication via environment variables
export OCI_USERNAME=myuser
export OCI_PASSWORD=mypassword
Port Configuration
The server can be configured to listen on a specific port using either:
-
Environment Variable:
MCP_PORT: The port number to listen on (must be between 0 and 65535)- If not set or invalid, defaults to port 8080
-
Command-line Flag:
-port: Overrides the environment variable setting (must be between 0 and 65535)- If invalid port provided it defaults to port 8080
- Example:
./ocireg-mcp -port 9090
Testing
go test ./...
Linting
golangci-lint run
Contributing
We welcome contributions to this MCP server! If you'd like to contribute, please review the CONTRIBUTING guide for details on how to get started.
If you run into a bug or have a feature request, please
open an issue in the
repository or join us in the #mcp-servers channel on our
community Discord server.
License
This project is licensed under the Apache v2 License - see the LICENSE file for details.
Star History
Repository Owner
Organization
Repository Details
Programming Languages
Tags
Join Our Newsletter
Stay updated with the latest AI tools, news, and offers by subscribing to our weekly newsletter.
Related MCPs
Discover similar Model Context Protocol servers
Docker Hub MCP Server
Expose Docker Hub APIs to LLMs via the Model Context Protocol.
The Docker Hub MCP Server implements the Model Context Protocol (MCP) to make Docker Hub APIs accessible to large language models, enabling AI-powered discovery and management of container images and repositories. It provides an interface for LLMs to access real-time Docker Hub data, recommend images, and streamline developer workflows. The server supports both public and private repositories through configurable authentication, and can be integrated with AI assistants like Gordon and clients such as Claude Desktop.
- ⭐ 83
- MCP
- docker/hub-mcp
OSV MCP Server
SSE-based MCP server for querying the OSV vulnerability database
OSV MCP Server implements a Model Context Protocol (MCP) server that provides secure, programmatic access to the Open Source Vulnerabilities (OSV) database. It supports real-time streaming communication using SSE or streamable HTTP and allows querying of vulnerabilities for single or multiple packages, versions, or commits. The server is designed for integration with LLM-powered applications and supports configuration through environment variables and containerized deployments via ToolHive.
- ⭐ 25
- MCP
- StacklokLabs/osv-mcp
Portainer MCP
Connect AI assistants securely to Portainer environments using the Model Context Protocol.
Portainer MCP is an implementation of the Model Context Protocol (MCP) designed for seamless integration between AI assistants and Portainer-managed container environments. It enables management of Portainer resources and allows execution of Docker and Kubernetes commands through AI interfaces in a secure, standardized manner. The tool provides direct protocol-based access to environment data, facilitating automation and operational insights for container infrastructures.
- ⭐ 81
- MCP
- portainer/portainer-mcp
FHIR MCP Server
A Model Context Protocol server for seamless interaction with FHIR resources and AI tools.
FHIR MCP Server implements a full Model Context Protocol server, enabling large language model agents to perform comprehensive CRUD operations on FHIR-compliant healthcare data. It offers standardized integration with various clinical data sources, natural-language query capabilities, and supports secure authentication via OAuth2. The server includes semantic search, AI-powered document processing, terminology resolution, Docker deployment, and is optimized for use with MCP-compatible clients like Claude Desktop.
- ⭐ 34
- MCP
- the-momentum/fhir-mcp-server
MCP Nutanix
An MCP server enabling LLM access to Nutanix Prism Central APIs.
MCP Nutanix is an experimental Model Context Protocol (MCP) server that allows large language models to interact with Nutanix Prism Central APIs. It facilitates listing and accessing resources such as VMs, clusters, and hosts via standardized MCP client-server integration, using the Prism Go Client for backend communication. The implementation supports both interactive and static credential methods, making it compatible with various MCP clients including Claude and Cursor.
- ⭐ 11
- MCP
- thunderboltsid/mcp-nutanix
OpsLevel MCP Server
Read-only MCP server for integrating OpsLevel data with AI tools.
OpsLevel MCP Server implements the Model Context Protocol to provide AI tools with a secure way to access and interact with OpsLevel account data. It supports read-only operations for a wide range of OpsLevel resources such as actions, campaigns, checks, components, documentation, domains, and more. The tool is compatible with popular environments including Claude Desktop and VS Code, enabling easy integration via configuration and API tokens. Installation options include Homebrew, Docker, and standalone binaries.
- ⭐ 8
- MCP
- OpsLevel/opslevel-mcp
Didn't find tool you were looking for?