← 返回 Skills 市场
Nuwa 女娲
作者
alchaincyf
· GitHub ↗
· v1.0.0
· MIT-0
1099
总下载
0
收藏
9
当前安装
1
版本数
在 OpenClaw 中安装
/install nuwa-skill
功能描述
Nuwa (女娲): Input any name, auto-research → extract thinking frameworks → generate a runnable perspective skill. Uses multi-agent parallel research, mental mo...
安全使用建议
This skill appears to do what it says: it runs parallel research agents, extracts mental models, and writes a SKILL.md that can answer in that perspective. Before installing, consider the following:
- Impersonation/legal risk: The examples and role rules instruct the generated perspective to speak as the named person (using 'I'). That raises impersonation and defamation risks for living people; confirm whether your platform policy permits such roleplay and ensure required disclaimers and safety controls are in place. Limit use to public figures or obtain consent when appropriate.
- Privacy and target scope: The skill accepts 'any name' — decide and enforce whether it should be restricted to public figures only. Researching private individuals could surface sensitive personal data.
- Copyright and paywalled sources: The workflow includes 'downloaded primary materials' and saving full research files. Confirm whether the agent will attempt to fetch paywalled or copyrighted content and whether that is acceptable; consider adding rules to avoid scraping paywalled content or to store only summaries/quotes within fair use limits.
- Data persistence & retention: The skill will create and retain a references/research directory with full MD files. If you don't want persistent copies, require the agent to store only ephemeral summaries or to clean up after generation.
- Triggering & safety controls: The SKILL.md includes many permissive activation triggers. Consider restricting triggers or requiring explicit user confirmation before harvesting data or creating new skill directories.
If you accept these trade-offs (and add guardrails: scope limits, disclaimers in generated skills, retention policies, and scraping/copyright rules), the skill is coherent with its stated purpose. If you need the skill to avoid impersonation, private targets, or persistent downloads, ask the author to modify SKILL.md accordingly before installing.
能力标签
能力评估
Purpose & Capability
Name/description align with the runtime instructions: the SKILL.md describes an automated research → extraction → skill-generation pipeline and the instructions implement that (parallel subagents, research files, template SKILL.md). The absence of required binaries, env vars, or installs is coherent for an instruction-only skill.
Instruction Scope
The instructions tell the agent to 'auto-research' any input name, spawn six parallel subagents, download and save primary materials into a local skill directory, and produce a SKILL.md that will role-play as the target (explicitly instructing using 'I' for real persons such as Elon Musk). That scope is broad: it can collect and persist large amounts of external content (including copyrighted or paywalled material), and it explicitly instructs impersonation-style roleplay for living public figures. The SKILL.md does not restrict targets to public figures or require consent, nor does it limit retention or specify safe sourcing for copyrighted/private material.
Install Mechanism
Instruction-only skill with no install spec, no packages pulled, no external installer URLs or archive extraction. This is low install risk — nothing will be written to disk by an installer beyond what the agent itself chooses to create at runtime per the instructions.
Credentials
The skill requests no environment variables, binaries, or credentials in metadata. The runtime instructions reference web research and saving files, but they don't ask for unrelated credentials or access to system config paths. Declared environment/credential footprint is minimal and proportionate to the stated purpose.
Persistence & Privilege
always:false and no special platform privileges. However, the instructions explicitly direct the agent to create a skill directory and persist full research outputs and 'downloaded primary materials' to disk. That means the agent will store potentially large amounts of scraped content persistently in its workspace; users should expect local persistence and verify retention/cleanup policies.
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 nuwa-skill - 安装完成后,直接呼叫该 Skill 的名称或使用
/nuwa-skill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 with Naval and Elon Musk examples
元数据
常见问题
Nuwa 女娲 是什么?
Nuwa (女娲): Input any name, auto-research → extract thinking frameworks → generate a runnable perspective skill. Uses multi-agent parallel research, mental mo... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1099 次。
如何安装 Nuwa 女娲?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nuwa-skill」即可一键安装,无需额外配置。
Nuwa 女娲 是免费的吗?
是的,Nuwa 女娲 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Nuwa 女娲 支持哪些平台?
Nuwa 女娲 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 Nuwa 女娲?
由 alchaincyf(@alchaincyf)开发并维护,当前版本 v1.0.0。
推荐 Skills