Skip to content
 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

dora-skills

English | 中文


English

A Claude Code skill package for building AI agents, workflows, and embodied intelligence with the dora-rs dataflow framework.

Focus: Embodied Intelligence — combining perception, language models, and physical control.

Quick Start

# Install dora CLI
pip install dora-rs

# Create a project
dora new my-robot --lang python

# Run a dataflow
dora run dataflow.yml

Commands

Command Description Documentation
/new-dataflow Create a new dataflow project docs
/add-node Add a node to existing dataflow docs
/build Build and install dependencies docs
/run Run dataflow in foreground docs
/start Start dataflow in daemon mode docs
/stop Stop running dataflow docs
/list List running dataflows docs
/visualize Generate dataflow visualization docs

Available Skills

Category Skills
Core core-development, custom-node, cli-workflow
Vision object-detection, segmentation, tracking, vlm
Audio speech-to-text, text-to-speech, voice-activity
Robot arm-control, actuators, chassis
Data recording, replay, lerobot

Using with Claude Code

Installation:

# Install Claude Code
npm install -g @anthropic-ai/claude-code

# Clone skill package to skills directory
git clone https://github.com/dora-rs/dora-skills.git ~/.claude-code/skills/dora-skills

Usage:

# Start Claude Code in your project
cd my-robot-project
claude-code

Claude will automatically use these skills when you ask about dataflow pipelines, vision/audio nodes, or robot control.

Using with Other AI Tools

Works with Cursor IDE, Continue.dev, GitHub Copilot, Aider, or any LLM via direct API integration. Reference the SKILL.md files as context.

Resources


中文

一个用于 Claude Code 的技能包,使用 dora-rs 数据流框架构建 AI Agent、工作流和具身智能应用。

重点: 具身智能 —— 融合感知、语言模型与物理控制。

快速开始

# 安装 dora CLI
pip install dora-rs

# 创建项目
dora new my-robot --lang python

# 运行数据流
dora run dataflow.yml

命令

命令 说明 文档
/new-dataflow 创建新的数据流项目 文档
/add-node 向现有数据流添加节点 文档
/build 构建和安装依赖 文档
/run 前台运行数据流 文档
/start 后台守护模式启动数据流 文档
/stop 停止运行中的数据流 文档
/list 列出运行中的数据流 文档
/visualize 生成数据流可视化图 文档

可用技能

类别 技能
核心 core-development, custom-node, cli-workflow
视觉 object-detection, segmentation, tracking, vlm
音频 speech-to-text, text-to-speech, voice-activity
机器人 arm-control, actuators, chassis
数据 recording, replay, lerobot

使用 Claude Code

安装:

# 安装 Claude Code
npm install -g @anthropic-ai/claude-code

# 克隆技能包到技能目录
git clone https://github.com/dora-rs/dora-skills.git ~/.claude-code/skills/dora-skills

使用:

# 在项目中启动 Claude Code
cd my-robot-project
claude-code

当你询问数据流管道、视觉/音频节点或机器人控制时,Claude 会自动使用这些技能。

使用其他 AI 工具

支持 Cursor IDE、Continue.dev、GitHub Copilot、Aider 或通过直接 API 集成使用任何 LLM。引用 SKILL.md 文件作为上下文即可。

资源


License

Apache-2.0

About

Skills for building AI agents, workflows, and embodied intelligence applications with the dora-rs dataflow framework

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages