VectorClaw MCP
VectorClaw connects OpenClaw to an Anki / Digital Dream Labs Vector robot through MCP.
It provides practical robot control primitives for speech, movement, camera capture, and status/sensor reads.
What you can do
- - Speak text with INLINECODE0
- Move and position with
vector_drive, vector_head, INLINECODE3 - Capture camera images with
vector_look and INLINECODE5 - Read robot state with
vector_status, vector_pose, vector_proximity_status, INLINECODE9 - Build look → reason → act workflows
Vision requirement for look → reason → act
For see → reason → act workflows, the agent must either be vision-capable itself (e.g., a VLM) or have access to a separate vision model/image-interpretation tool to analyze camera images before choosing actions.
Requirements
- - Vector robot configured and reachable
- Wire-Pod running
- SDK configured at INLINECODE10
- INLINECODE11 environment variable set
Quick setup
- 1. Install package: INLINECODE12
- Configure SDK: INLINECODE13
- Export robot serial: INLINECODE14
- Add MCP server:
CODEBLOCK0
Tool coverage
Hardware-verified core tools
vector_say, vector_drive_off_charger, vector_drive, vector_emergency_stop, vector_head, vector_lift, vector_look, vector_capture_image, vector_face, vector_scan, vector_vision_reset, vector_pose, vector_status, vector_charger_status, vector_touch_status, INLINECODE30
Experimental tools
vector_animate, vector_drive_on_charger, vector_find_faces, vector_list_visible_faces, vector_face_detection, vector_list_visible_objects, INLINECODE37
Current limitations
- - Charger return (
vector_drive_on_charger) is currently unreliable - Face/object detection is currently inconsistent
- Visual interpretation requires the vision capability described above
Documentation
- - MCP API: INLINECODE39
- SDK Reference: INLINECODE40
- Hardware log: INLINECODE41
- Repo: https://github.com/danmartinez78/VectorClaw
VectorClaw MCP
VectorClaw 通过 MCP 将 OpenClaw 连接到 Anki / Digital Dream Labs Vector 机器人。
它提供了实用的机器人控制原语，用于语音、移动、摄像头捕捉以及状态/传感器读取。
你能做什么
- - 使用 vectorsay 说出文本
- 使用 vectordrive、vectorhead、vectorlift 进行移动和定位
- 使用 vectorlook 和 vectorcaptureimage 捕捉摄像头图像
- 使用 vectorstatus、vectorpose、vectorproximitystatus、vectortouch_status 读取机器人状态
- 构建观察→推理→行动的工作流程
观察→推理→行动的视觉要求
对于看→推理→行动的工作流程，智能体本身必须具有视觉能力（例如，VLM），或者能够访问单独的视觉模型/图像解释工具，以便在选择行动前分析摄像头图像。
要求
- - Vector 机器人已配置且可访问
- Wire-Pod 正在运行
- SDK 配置在 ~/.ankivector/sdkconfig.ini
- 已设置 VECTOR_SERIAL 环境变量
快速设置
- 1. 安装包：pip install vectorclaw-mcp
- 配置 SDK：python3 -m ankivector.configure
- 导出机器人序列号：export VECTORSERIAL=your-serial
- 添加 MCP 服务器：
json
{
mcpServers: {
vectorclaw: {
command: python3,
args: [-m, vectorclaw_mcp.server],
env: { VECTORSERIAL: ${VECTORSERIAL} }
}
}
}
工具覆盖范围
硬件验证的核心工具
vectorsay、vectordriveoffcharger、vectordrive、vectoremergencystop、vectorhead、vectorlift、vectorlook、vectorcaptureimage、vectorface、vectorscan、vectorvisionreset、vectorpose、vectorstatus、vectorchargerstatus、vectortouchstatus、vectorproximitystatus
实验性工具
vectoranimate、vectordriveoncharger、vectorfindfaces、vectorlistvisiblefaces、vectorfacedetection、vectorlistvisibleobjects、vector_cube
当前限制
- - 返回充电器（vectordriveon_charger）目前不可靠
- 面部/物体检测目前不稳定
- 视觉解释需要上述的视觉能力
文档
- - MCP API：docs/MCPAPIREFERENCE.md
- SDK 参考：docs/VECTORSDKREFERENCE.md
- 硬件日志：docs/HARDWARESMOKELOG.md
- 仓库：https://github.com/danmartinez78/VectorClaw
