DoorDash has launched a limited beta version of its AI command-line tool, DoorDash CLI (dd-cli), which allows developers to complete food delivery orders directly through an AI assistant, including searching for merchants, finding promotions, and completing checkout. The tool is currently available to macOS developers in the United States and Canada through a waitlist.

Andy Fang, co-founder and CTO of DoorDash, announced this new feature on the X platform. While command-line tools are typically associated with software development scenarios, using them for ordering food may seem like a "programmer's joke." However, the DoorDash CLI actually represents an exploration of agentic commerce.

By opening up its meal ordering capabilities, DoorDash hopes that AI agents can call its service APIs, allowing developers to integrate food ordering, grocery purchasing, and local promotion inquiries into their own software, smart assistants, or other services without entering the DoorDash app. In the future, AI agents could become a new entry point connecting consumers with commercial services.

Delivery, Ordering, Cooking Tutorial (2)

Previously, DoorDash had provided meal ordering services via iMessage and launched an AI chatbot called "Ask DoorDash." It also opened up related capabilities to third-party AI assistants such as OpenAI ChatGPT and Claude, continuing to build an AI-driven consumer service ecosystem.

The current dd-cli test also demonstrates how AI agents can be applied in real tasks. In a demonstration video, the AI assistant reads Slack messages, parses JSON data, checks menu structures, runs Python scripts, and adjusts operations based on error feedback, ultimately completing multiple salad orders. Although the process is much more complex than regular ordering, it reflects the direction of AI agents autonomously calling tools and executing tasks.

As AI assistants gradually move from information interaction to task execution, companies like DoorDash are trying to directly integrate commercial service capabilities into the AI ecosystem, driving the transformation of consumer scenarios from "users actively operating apps" to "AI agents handling services on behalf of users."