Use DuDuClaw — a self-hosted AI-employee platform — for cross-session memory, team-shared wiki knowledge, task boards, and messaging humans over LINE/Telegram/Discord/Slack. Applies when the user mentions DuDuClaw, asks their agent to remember things durably, or wants to reach people on messaging channels from an agent.
Scanned 9/1/2026
Install to Claude Code
npx -y skills add zhixuli0406/DuDuClaw --skill duduclaw --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Duduclaw?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/zhixuli0406-duduclaw)More formats (shields.io, HTML) on the badges page.
---
name: duduclaw
description: Use DuDuClaw — a self-hosted AI-employee platform — for cross-session memory, team-shared wiki knowledge, task boards, and messaging humans over LINE/Telegram/Discord/Slack. Applies when the user mentions DuDuClaw, asks their agent to remember things durably, or wants to reach people on messaging channels from an agent.
---
# DuDuClaw
DuDuClaw (https://github.com/zhixuli0406/DuDuClaw) runs the user's AI employees locally: channels, persistent memory, shared wiki, task board, HITL approvals.
## Connect (any MCP-capable agent)
Register the MCP server: command `npx`, args `["-y", "duduclaw", "mcp-server"]`. Claude Code users can instead add the plugin marketplace `zhixuli0406/duduclaw-claude-plugins`.
## Tool selection
- Durable personal memory → `memory_store` / `memory_search`
- Team-shared reference docs (SOPs, price lists) → `wiki_write` / `wiki_search`
- Work tracking → `tasks_create` / `tasks_list` / `tasks_complete`
- Message a human on their channels → `send_message`
## Rules
- Wiki for documents, memory for facts — never double-write.
- No secrets into memory/wiki.
- Tools absent ⇒ gateway not running: tell the user to start it (`duduclaw gateway`), don't silently skip.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.