Building dynamic, intelligent, and automated systems today often involves creating "Agentic Workflows" and running "Services-as-Software." These aren't just theoretical concepts; they represent the future of "business-as-code," where complex processes are orchestrated and executed autonomously or semi-autonomously.
But bringing these concepts to life requires a solid foundation and clear guidance. How do you define agent interactions? What do workflow outputs look like? How do you integrate external services seamlessly? This is where frustration can set in without a definitive guide.
Introducing reference.do, the Canonical Reference for Agentic Flows on the .do platform.
Reference.do is designed to be the single, comprehensive source of truth for developers, architects, and business users looking to understand and build on the .do platform. It provides detailed .do documentation covering every aspect you need to transform business logic into executable, reliable code.
Whether you're just starting to explore the possibilities of agentic API integrations or you're deeply involved in architecting sophisticated services-as-software API structures, reference.do offers the clarity required.
What exactly can you expect to find within this agentic workflow reference? The documentation is meticulously structured to provide actionable insights and technical specifications:
The backbone of interacting with any platform is its API. Reference.do provides extensive API documentation, detailing available endpoints, required parameters, request and response formats, authentication methods, and error handling. This ensures you can confidently connect your applications and services to the .do platform.
To build on a platform, you must understand its fundamental building blocks. The reference includes clear explanations and schemas for the platform's core data models. Learn the structure of Agents, the flow definition of Workflows, the input/output schemas for Services, and how different components interact. This knowledge is crucial for correctly structuring your own agents and services.
For instance, understanding the defined structure for an AgentStepOutput or the overall AgentWorkflowOutput (like the TypeScript examples found in the reference) is vital for processing results and building robust workflows:
These clear definitions remove ambiguity and speed up development.
Documentation is powerful, but seeing is believing. Reference.do is populated with practical code examples demonstrating how to perform common tasks, implement specific patterns, and integrate with various parts of the platform using the services-as-software API and workflow APIs. These examples serve as accelerators for your own development process.
Beyond just the raw technical specs, the reference also includes guides on architecting effective Agentic Workflows and building performant, scalable Services-as-Software. Learn best practices for designing agents, orchestrating steps, managing state, and ensuring the reliability of your "business-as-code" initiatives.
Reference.do is more than just static documentation; it's the bridge that connects your ideas for automated business processes to their functional implementation on the .do platform. It's the essential tool for anyone serious about leveraging Agentic Flows and Services-as-Software.
Ready to start building? Explore the reference.do - Canonical Reference for Agentic Flows today and transform your business concepts into executable code.
interface AgentStepOutput {
status: 'success' | 'failure';
output: any; // The output of the agent step can be anything
agentId: string; // The ID of the agent that executed this step
}
interface AgentWorkflowOutput {
status: 'completed' | 'failed';
steps: AgentStepOutput[];
finalOutput: any; // The final output of the entire workflow
}