← 返回 Skills 市场
5579
总下载
3
收藏
44
当前安装
1
版本数
在 OpenClaw 中安装
/install computer-vision-expert
功能描述
SOTA Computer Vision Expert (2026). Specialized in YOLO26, Segment Anything 3 (SAM 3), Vision Language Models, and real-time spatial analysis.
安全使用建议
Reasonable to install for computer vision guidance. Users should still verify time-sensitive technical claims such as specific 2026 model names, benchmark claims, and deployment recommendations before relying on them in production.
功能分析
Type: OpenClaw Skill
Name: computer-vision-expert
Version: 1.0.0
The skill bundle consists of a metadata file and a markdown file describing the capabilities of a 'Computer Vision Expert' agent. The content of SKILL.md is purely descriptive, outlining expertise in various computer vision domains (YOLO26, SAM 3, VLMs, geometry). There are no instructions that suggest prompt injection, data exfiltration, malicious execution, persistence, or any other high-risk behaviors. The files are clearly aligned with the stated purpose of defining an AI agent's skill set.
能力评估
Purpose & Capability
The SKILL.md content matches the stated purpose: it gives computer vision architecture and model-selection guidance for detection, segmentation, VLMs, edge deployment, and geometry.
Instruction Scope
The activation language is broad across computer vision topics, but it remains domain-specific and does not request elevated authority, override user control, or introduce unrelated behavior.
Install Mechanism
The artifact contains only a markdown skill file and metadata context; there are no executable scripts, package installation steps, remote endpoints, or setup commands.
Credentials
The skill does not request filesystem, network, device, credential, account, profile, or private data access.
Persistence & Privilege
No persistence, background workers, automatic execution, privilege escalation, memory storage, or mutation authority is present.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install computer-vision-expert - 安装完成后，直接呼叫该 Skill 的名称或使用
/computer-vision-expert触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of Computer Vision Expert (SOTA 2026):
- Provides expert guidance on state-of-the-art computer vision, including YOLO26, SAM 3, and Vision Language Models.
- Covers real-time detection, promptable segmentation, and advanced spatial/3D reasoning.
- Includes best practices for edge deployment, text-guided vision pipelines, and geometry-based reconstruction.
- Documents common anti-patterns and sharp-edge issues for 2026-era vision models.
元数据
常见问题
Computer Vision Expert 是什么？
SOTA Computer Vision Expert (2026). Specialized in YOLO26, Segment Anything 3 (SAM 3), Vision Language Models, and real-time spatial analysis. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 5579 次。
如何安装 Computer Vision Expert？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install computer-vision-expert」即可一键安装，无需额外配置。
Computer Vision Expert 是免费的吗？
是的，Computer Vision Expert 完全免费（开源免费），可自由下载、安装和使用。
Computer Vision Expert 支持哪些平台？
Computer Vision Expert 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Computer Vision Expert？
由 Zorro Ng（@zorrong）开发并维护，当前版本 v1.0.0。
推荐 Skills