← 返回 Skills 市场
646
总下载
0
收藏
8
当前安装
1
版本数
在 OpenClaw 中安装
/install numpy
功能描述
Write fast, memory-efficient numerical code with arrays, broadcasting, vectorization, and linear algebra.
安全使用建议
This skill appears coherent and local: it needs python3 and will create ~/numpy/ to store preferences and code snippets and will write a preference into the agent's MAIN memory to remember when to activate. Before installing, confirm you’re comfortable with the skill saving code/snippets locally and adding activation preferences to the agent memory. Also ensure python3 is the intended interpreter on your system. There is no declared network activity or secret access, but remember the SKILL.md is a set of instructions the agent will follow — if you want stronger guarantees, inspect agent logs or the created ~/numpy/ files after first use.
功能分析
Type: OpenClaw Skill
Name: numpy
Version: 1.0.0
The OpenClaw NumPy skill bundle is classified as benign. All files (SKILL.md, memory-template.md, setup.md) consistently describe a skill focused on assisting with NumPy operations, storing user preferences and code snippets locally within `~/numpy/`. The `SKILL.md` explicitly states that the skill does NOT send data externally, access files outside `~/numpy/`, or require network connectivity, and no other instructions contradict this. There is no evidence of prompt injection attempts, data exfiltration, persistence mechanisms, or other malicious behaviors.
能力评估
Purpose & Capability
Name/description (NumPy helper) match the declared binary requirement (python3) and the content of SKILL.md: teaching and saving NumPy patterns and preferences. No extraneous credentials, config paths, or unrelated binaries are requested.
Instruction Scope
SKILL.md instructs the agent to create and read/write under ~/numpy/ (memory.md and snippets). That is coherent for a snippet/preference helper. It also mentions saving a preference to the agent's MAIN memory so the skill knows when to activate; this is a reasonable behavior but does extend beyond the skill's own folder into agent-level memory.
Install Mechanism
Instruction-only skill with no install spec and no code files — minimal risk because nothing is downloaded or written by an installer. It relies on an existing python3 binary, which is appropriate.
Credentials
No environment variables, secrets, or unrelated credentials are requested. Requested filesystem access is limited to a user-owned directory (~/numpy/) as documented.
Persistence & Privilege
always:false (normal) and disable-model-invocation:false (normal). The skill will persist data under ~/numpy/ and writes some preferences into MAIN memory per setup.md; persisting user preferences/snippets is expected, but users should be aware it will store code and preferences locally and record activation choices in agent memory.
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 numpy - 安装完成后，直接呼叫该 Skill 的名称或使用
/numpy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
元数据
常见问题
NumPy 是什么？
Write fast, memory-efficient numerical code with arrays, broadcasting, vectorization, and linear algebra. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 646 次。
如何安装 NumPy？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install numpy」即可一键安装，无需额外配置。
NumPy 是免费的吗？
是的，NumPy 完全免费（开源免费），可自由下载、安装和使用。
NumPy 支持哪些平台？
NumPy 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（linux, darwin, win32）。
谁开发了 NumPy？
由 Iván（@ivangdavila）开发并维护，当前版本 v1.0.0。
推荐 Skills