The Open Digital Architecture (ODA) Canvas is an execution platform for ODA Components. This repository contains the Reference Implementation, open-source code, use cases and test-kit for a TM Forum ODA Canvas. You are free to use this as a starting point for your own ODA Canvas implementation. Explore available ODA Components in the ODA Component Directory. The Reference Implementation of the ODA Canvas will be used for ODA Component certification.

see Installation Guide for installation instructions.

There is an introductory video on the ODA Canvas at:

The ODA Canvas is itself a modular architecture, with independent operators that embed the management and operations activities in software. The Component Management operator manages the lifecycle of each component, and the decomposition into ExposedAPIs, IdentityConfigs and other sub-resources (that are processed by their corresponding operators). The API Management operators manage the API Gateway and/or Service Mesh to provide security, throttling and other non-functional services to allow API endpoints to be exposed to external consumers. You deploy the matching API Management operator for whatever API Gateway or Service Mesh you are using. The Identity Config operator configures the Identity Management Services. The Secrets Management operator is an optional operator to configure secrets. The Dependency Management operator provides services to allow a Component to discover API dependencies.

The Canvas now includes support for AI-Native capabilities, enabling ODA Components to expose MCP (Model Context Protocol) interfaces as tools for AI agents, and allowing AI agents themselves to be deployed as composable ODA Components. This includes AI-specific services such as AI Gateways for responsible AI practices, observability for language model interactions, and continuous evaluation frameworks for non-deterministic AI systems. For more information, see AI-Native Canvas Design.

We foresee new operator types becoming available as the ODA Canvas matures. The ODA Canvas reference implementation includes example operators to that are freely available for organizations to re-use, extend or replace with their own implementations. We expect a typical production implementation will use a combination of standard operators and custom operators that can implement that organizations specific operational policies.

For more information see Software Operators

The ODA Canvas design is to allow ODA Components to declare their requirements for underlying capabilities in a technology-independent way. The Canvas manages these underlying capabilities using Software Operators that follow the Kubernetes Operator Pattern.

The design (including ongoing work) of the ODA Canvas is documented in ODA Canvas Design. The Use Case Library shows how ODA Components interact with the ODA Canvas. The Feature definition and Test Kit details the features required to be a fully compliant ODA Canvas and a set of tests that can be used to validate any ODA Canvas. The source code contains the source code for the operators that are part of the ODA Canvas.

Security principles for the ODA Canvas design are documented in ODA Security Principles.

How to work with this repository:

• How-To work with Dockerimages