DoorDash opens limited beta for command-line tool targeting AI agents
The US-based delivery giant is expanding its agentic commerce strategy, offering macOS developers in North America early access to a tool that enables software to interact with its platform without human intervention.
DoorDash has launched a limited beta of dd-cli, a command-line interface tool designed to allow developers and artificial intelligence agents to interact directly with its delivery platform. The tool enables users to search for stores, identify deals, build shopping carts, and place orders from a terminal environment, marking a strategic shift toward agentic commerce where software agents autonomously perform tasks previously reserved for human users via graphical interfaces.
The beta is currently restricted to developers using macOS operating systems located in the United States and Canada, with access managed via a waitlist. DoorDash co-founder and chief technology officer Andy Fang announced the launch via a post on X, highlighting the tool’s capability to parse JSON, run Python scripts, and recover from errors. The announcement included a demonstration video that framed the technology with references to programming culture, illustrating how the system can handle complex tasks such as reading Slack messages and calculating totals to order multiple items.
This initiative represents a move toward agentic commerce, a model where AI agents act on behalf of users to complete transactions across digital platforms. By exposing its ordering platform to developers, DoorDash is enabling the creation of third-party tools that can automate food, grocery, and local deal searches. The company’s sign-up form for the beta includes a field asking developers to specify what they intend to build, underscoring the platform’s role as a building block for broader software integrations.
DoorDash has previously experimented with this approach through its integration with Apple’s iMessage and its own AI chatbot, Ask DoorDash. The company also exposes its services to third-party AI chatbots, including OpenAI’s ChatGPT and Anthropic’s Claude. The launch of dd-cli extends this strategy by allowing direct programmatic access to its core ordering functions, rather than relying solely on conversational interfaces or messaging apps.
While the concept of ordering lunch via command line may appear humorous, the underlying technology signals a serious effort to embed DoorDash’s services into diverse digital ecosystems. The tool allows developers to add functionality to their own software and services, potentially automating mundane tasks that were traditionally performed manually. As the company continues to refine its agentic commerce capabilities, the dd-cli beta serves as an early indicator of how consumer platforms may evolve to accommodate autonomous software agents.
