← 返回 Skills 市场
1616
总下载
3
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install skill-mixer
功能描述
Processes and auto-categorizes 150 skills into 12 composite master skills with ClawHub integration and multiple deployment options.
安全使用建议
Do not follow the build/upload commands or supply credentials until the package contents and origin are clarified. The SKILL.md claims many source files and docs that are not present in the registry entry — this could be an accidental incomplete publish or an attempt to get you to fetch/run external code or reveal secrets. Ask the publisher for: (1) a link to the authoritative repository or release, (2) the missing files (package.json, source, docs), and (3) an explicit list of any required environment variables and endpoints. If you decide to run build/upload steps later, first inspect the full repository locally, review package.json and all code, and run builds inside an isolated VM or container. If you already provided credentials after following these instructions, rotate them and audit any activity.
功能分析
Type: OpenClaw Skill
Name: skill-mixer
Version: 1.0.0
The provided `_meta.json` and `SKILL.md` files contain standard metadata and documentation for a skill bundle. The `SKILL.md` serves as a comprehensive guide for a human user, detailing the skill's purpose, features, and steps for local testing, building, and publishing to ClawHub. There are no instructions within `SKILL.md` that attempt prompt injection against an AI agent, nor any commands or directives that suggest malicious intent, data exfiltration, or unauthorized actions. The commands mentioned (e.g., `pnpm cli`, `pnpm build`) are standard development operations, and the instructions are clearly directed at the user for managing the skill.
能力评估
Purpose & Capability
The runtime instructions describe a 26-file TypeScript project (source files, adapters, build steps, deployment guides) and recommend building and uploading to clawhub.ai, but the registry entry contains only SKILL.md and no code files, package.json, or homepage. The declared metadata contains no description or repository link. A user expecting a packaged SkillMixer project would legitimately need the listed source and packaging files — their absence is an incoherence.
Instruction Scope
SKILL.md instructs the operator/agent to read many local docs (START_HERE.md, CLAWHUB_PUBLISH_GUIDE.md, etc.), run pnpm build or pnpm --filter ... cli, and "configure credentials" for deployment. None of those files exist in the bundle and no specific credential names or endpoints are declared. The instructions are actionable (run build/upload, configure credentials) but refer to artifacts and secrets that are not provided nor enumerated, which could prompt the agent or user to fetch external content or reveal credentials.
Install Mechanism
There is no install spec and no code files — from an installation-risk perspective this is lower risk because nothing is written or executed by the skill itself. However, the SKILL.md expects a project build and external tooling (pnpm) that would execute if the user follows the instructions; the lack of a package means a user might be encouraged to fetch or run unknown code from elsewhere, which shifts the risk to out-of-band actions.
Credentials
The registry lists no required environment variables or primary credential, yet the instructions tell users to "Configure credentials" for cloud/K8s/ClawHub deployment without specifying what credentials are needed. This omission is disproportionate: a deployable ClawHub adapter would normally declare which API keys or tokens it needs. The vagueness raises the risk the user or agent may be asked to supply unrelated secrets later.
Persistence & Privilege
The skill does not request always:true, does not ship code or install hooks, and does not declare any config-path or system modifications. Autonomous model invocation is allowed by default (normal) but is not combined here with any declared broad privileges.
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 skill-mixer - 安装完成后,直接呼叫该 Skill 的名称或使用
/skill-mixer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
SkillMixer v1.0.0 – Initial Release
- Production-ready clustering skill with built-in ClawHub integration.
- Auto-categorizes up to 150 skills into 12 master categories.
- Zero dependencies; pure TypeScript with error handling and tests.
- Comprehensive documentation (14 guides) covering usage, deployment, and integration.
- Multiple deployment and usage options: ClawHub upload, cloud/self-host, and local integration.
- Includes configuration templates and quickstart navigation.
元数据
常见问题
Skill Mixer 是什么?
Processes and auto-categorizes 150 skills into 12 composite master skills with ClawHub integration and multiple deployment options. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1616 次。
如何安装 Skill Mixer?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install skill-mixer」即可一键安装,无需额外配置。
Skill Mixer 是免费的吗?
是的,Skill Mixer 完全免费(开源免费),可自由下载、安装和使用。
Skill Mixer 支持哪些平台?
Skill Mixer 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 Skill Mixer?
由 grxkun(@grxkun)开发并维护,当前版本 v1.0.0。
推荐 Skills