← 返回 Skills 市场
VMamba Env Doctor
作者
jiangkaiqi2005
· GitHub ↗
· v1.0.0
· MIT-0
88
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install vmamba-env-doctor
功能描述
Guide for setting up a stable VMamba environment and resolving common CUDA, PyTorch, MMCV, and compilation errors. Use when users encounter issues like "unsu...
安全使用建议
This skill is a coherent, copyable setup guide for VMamba; it does not request credentials or hide endpoints. Before running commands, review them and run them in an isolated conda environment. Confirm your system GPU drivers support CUDA 11.8, avoid running commands as root unless necessary, and inspect any local code you build (the pip install . step in kernels/selective_scan compiles native code). Be aware that mim/pip/conda will fetch packages from public repositories—ensure your environment/trust policies allow those sources.
功能分析
Type: OpenClaw Skill
Name: vmamba-env-doctor
Version: 1.0.0
The skill provides legitimate instructions and shell commands for setting up a VMamba development environment, including specific versions of PyTorch, CUDA, and OpenMMLab libraries. The commands in SKILL.md and README.md are standard package management operations (conda, pip, mim) and do not exhibit any signs of data exfiltration, malicious execution, or prompt injection.
能力评估
Purpose & Capability
The name/description match the content: SKILL.md is a step-by-step environment and build guide for VMamba, PyTorch, MMCV and compiling selective_scan. All requested actions (conda/pip/mim installs, setting CC/CXX, building a local kernel) are expected for that purpose.
Instruction Scope
Instructions are narrowly scoped to creating a conda env, installing specific package versions, fixing MKL/NumPy/OpenCV issues, and compiling a local CUDA kernel. The skill does not instruct reading unrelated files, exporting secrets, or contacting unexpected endpoints; network activity is limited to conda/pip/mim package repositories (expected).
Install Mechanism
This is instruction-only (no install spec). The guide tells users to run conda/pip/mim commands which are standard package managers. Nothing is downloaded from obscure personal URLs by the skill itself.
Credentials
No environment variables, credentials, or config paths are required by the skill. The guidance suggests setting CC/CXX for local compilation (expected and limited in scope).
Persistence & Privilege
always:false and no persistent modifications to agent/system configuration are requested by the skill. It does not ask to enable itself or modify other skills.
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 vmamba-env-doctor - 安装完成后，直接呼叫该 Skill 的名称或使用
/vmamba-env-doctor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 vmamba-env-doctor.
- Provides a step-by-step guide for setting up a stable VMamba environment.
- Covers solutions for common errors including CUDA, PyTorch, MMCV, and compilation issues.
- Includes tested environment versions and complete installation commands.
- Offers troubleshooting for errors like "unsupported GNU version", "libc10.so not found", and package conflicts.
元数据
常见问题
VMamba Env Doctor 是什么？
Guide for setting up a stable VMamba environment and resolving common CUDA, PyTorch, MMCV, and compilation errors. Use when users encounter issues like "unsu... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 88 次。
如何安装 VMamba Env Doctor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install vmamba-env-doctor」即可一键安装，无需额外配置。
VMamba Env Doctor 是免费的吗？
是的，VMamba Env Doctor 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
VMamba Env Doctor 支持哪些平台？
VMamba Env Doctor 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 VMamba Env Doctor？
由 jiangkaiqi2005（@jiangkaiqi2005）开发并维护，当前版本 v1.0.0。
推荐 Skills