Developers, DingTalk CLI is now open source—come discuss on GitHub?

DingTalk CLI is now open source! On March 27, the DingTalk CLI open-source project was officially launched on the GitHub community under the Apache-2.0 license. The first release opens up ten core product features: AI spreadsheets, calendar, logs, to-do lists, bots, contacts, DING messages, attendance, Open Platform documentation, and workspace—natively supporting mainstream AI coding and Agent execution environments such as Claude Code and Cursor. This marks the first national-level application in China to fully open all its product capabilities to AI Agents.

Lowering the Barrier for AI Agent Integration, Accelerating Enterprise Intelligence

Whether AI Agents can truly "get things done" within enterprise environments has become a key factor in the next phase of AI competition. In the past, developers and AI Agents had to read hundreds of pages of API documentation, write authentication logic, and manage complex authorization processes, resulting in average integration cycles lasting several weeks—severely limiting deployment efficiency.

On March 17, Alibaba launched “Wukong,” the world’s first enterprise-grade AI-native workspace, announcing that DingTalk had completed a full CLI (command-line interface)重构, rewriting its underlying codebase for AI so that all product functions can be natively invoked by AI. This transformation forms the technical foundation enabling Wukong to “take action.”

Ten days later, DingTalk officially open-sourced this CLI functionality. Developers and AI Agents no longer need to understand low-level API details—enterprise-grade features can now be accessed with just one command. The project is already live on GitHub; search for "dingtalk-workspace-cli" to access it. Developers and enterprises can join the official community via the README and participate in co-building new features.

Covering Complete Workflows, Supporting Commercial Use and Self-Hosting

The initial release of DingTalk CLI covers the entire enterprise collaboration workflow: AI spreadsheets enable automatic input, querying, and updating of business data; the calendar module can check team members’ availability and automatically schedule meetings; bots and DING messages support sending structured notifications and high-priority alerts to groups or individuals; the contact directory helps AI quickly locate organizational owners; to-do and log modules enable automated task creation, progress tracking, and closed-loop automation for daily and weekly reports.

All features are released under the Apache-2.0 license, allowing commercial use, code auditing, and permitting enterprises to self-host and modify the software internally without needing individual permissions—significantly reducing procurement decision barriers and compliance costs.

Natively Integrated with Mainstream AI Environments, Architecture Centered on AI

From the outset, DingTalk CLI was designed with AI as the primary use case, offering native support for leading AI coding and Agent execution environments such as Qoder, Claude Code, and Cursor. After a single initialization setup, developers can directly invoke DingTalk functions within any CLI-compatible Agent framework—without requiring additional SDK integration or middleware development.

Strategic Significance Behind Open-Sourcing: Competing for Infrastructure Definition in the AI Era

This open-sourcing initiative is more than just a tool release—it's a strategic move. Within the global AI capability stack, large models solve the question of “thinking,” while enterprise-grade Action capabilities—enabling AI Agents to “do”—are the rarest and most strategically valuable components.

The open-sourcing of DingTalk CLI marks the first time Alibaba’s ATH division has publicly showcased its developer ecosystem strategy since its formation, signaling a significant step in building enterprise infrastructure for the AI era. By going open-source, DingTalk actively participates in shaping the future platform landscape.

Zhu Hong, CTO of DingTalk, said: “We want every AI Agent to be able to call upon DingTalk as naturally as invoking a system command. Open-sourcing is our most direct commitment to the global developer community—and a clear statement about what enterprise infrastructure in the AI Agent era should look like.”

We dedicated to serving clients with professional DingTalk solutions. If you'd like to learn more about DingTalk platform applications, feel free to contact our online customer service or email at This email address is being protected from spambots. You need JavaScript enabled to view it.. With a skilled development and operations team and extensive market experience, we’re ready to deliver expert DingTalk services and solutions tailored to your needs!