</>MCP Agents Market
Skill

dingtalk-workspace-cli

by DingTalk-Real-AI2.9kGoUpdated 2026-09-04

DingTalk Workspace is an officially open-sourced cross-platform CLI tool from DingTalk. It unifies DingTalk’s full suite of product capabilities into a single package, is designed for both human users and AI agent scenarios.

Claude CodeClaude DesktopCursorCodexWindsurf

DingTalk Workspace CLI (dws) is an official cross-platform command-line tool that exposes DingTalk's full enterprise product suite as agent skills for AI assistants. It provides structured JSON responses and ready-to-use SKILL.md files that enable AI agents like Claude Code, Cursor, and Codex to manage contacts, calendars, tasks, documents, chat, approvals, and 20+ other DingTalk services. The CLI features OAuth device-flow authentication, zero-trust security with encrypted token storage, and automatic AI input correction for common model mistakes. Available in both per-product multi-skill and all-in-one mono-skill layouts, it's designed for both human terminal use and AI agent automation.

Key Features

20+ DingTalk services exposed as agent skills: contacts, chat, calendar, todo, approvals, AI tables, docs, whiteboards, drive, meeting minutes, mail, and more
Multi-skill and mono-skill layouts: per-product skills (dingtalk-calendar, dingtalk-chat) or single unified skill, installable via npx skills add or dws skill setup
AI-native design: auto-corrects common model mistakes (camelCase→kebab-case, sticky arguments, fuzzy matching), structured JSON output, jq filtering
Real-time event subscription: dws event consume streams DingTalk events (messages, @mentions, approvals, todos, card callbacks) as NDJSON for event-driven agents
Schema introspection: dws schema provides command discovery and execution contracts for agent tool selection
Zero-trust security: OAuth device-flow auth, PBKDF2+AES-256-GCM encrypted tokens, domain allowlist, dry-run preview, full audit trail through DingTalk Open Platform
Raw API access: dws api calls App Token-capable DingTalk OpenAPIs directly with automatic token management and auto-pagination
DingTalk bot bridge: dws dev connect links a DingTalk robot to local AI CLI (Claude Code, Codex) with multi-turn memory

Use Cases

  • 01AI agents creating calendar events, finding meeting rooms, and scheduling across multiple attendees
  • 02Automated todo and task management: creating, listing, and completing tasks with priorities and executors
  • 03Contact and organization search: finding users, departments, and roster profiles by name, role, or department
  • 04Document and knowledge management: searching, reading, and editing DingTalk docs, whiteboards, and wiki pages
  • 05Chat automation: sending messages, managing groups, and subscribing to @mentions or specified senders for event-driven workflows
  • 06Approval workflows: querying pending tasks, approving/rejecting OA instances, and monitoring approval events

Related Skills

View more

superpowers

An agentic skills framework & software development methodology that works.

Skill293.5k

skills

Skills for Real Engineers. Straight from my .agents directory.

Skill273k

skills

Public repository for Agent Skills

Skill179.2k

ponytail

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

Skill149.2k

dingtalk-workspace-cli — FAQ

What is DingTalk Workspace CLI?+

DingTalk Workspace CLI (dws) is an official cross-platform command-line tool from DingTalk that exposes 20+ enterprise services (chat, calendar, todos, docs, approvals) as structured agent skills for AI assistants. It provides both human-friendly terminal commands and AI-optimized JSON responses with SKILL.md files for tool integration.

How do I install DingTalk Workspace CLI agent skills?+

Install via npx skills add DingTalk-Real-AI/dingtalk-workspace-cli -g for automatic per-product skill installation, or use dws skill setup for interactive mode selection (multi or mono layout). The installer prompts for confirmation before writing to agent directories like ~/.cursor/skills or ~/.claude/skills.

Which AI clients and agents does it work with?+

Works with Claude Code, Cursor, Codex, Claude Desktop, Windsurf, ZCode, OpenCode, and Qoder. The skills are Markdown-based (SKILL.md) and install into standard agent skill directories recognized by these clients.

What are the prerequisites and authentication requirements?+

Requires DingTalk enterprise account with admin authorization for CLI access. Run dws auth login for OAuth device-flow authentication (browser opens automatically). For headless environments use dws auth login --device. Custom apps require Client ID and Client Secret for ISV/CI integration.

Is DingTalk Workspace CLI free and open source?+

Yes, it's Apache-2.0 licensed and fully open source. The project is in co-creation phase and accesses DingTalk enterprise data, requiring enterprise admin authorization for access.

How do I use it with AI agents after installation?+

After installing skills, AI agents can invoke DingTalk commands via natural language. Use --yes to skip confirmation prompts for agents, --dry-run to preview operations safely, and --jq to extract precise fields. Commands return structured JSON with schemas available via dws schema <command> --compact.

How do I install dingtalk-workspace-cli?+

Open the source repository on GitHub and follow its README. dingtalk-workspace-cli is a skill — MCP Agents Market links you directly to the official repo.

Is dingtalk-workspace-cli free?+

dingtalk-workspace-cli is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches

Productivity & Workflow SkillsCommunication SkillsDeveloper Tools SkillsBest MCP ServersBest PluginsBest AgentsTop rated listingsClaude Code toolsBrowse all categories