← 返回 Skills 市场
wangm-a3

Openclaw Enterprise

作者 WangM-A3 · GitHub ↗ · v1.2.3 · MIT-0
cross-platform ⚠ suspicious
233
总下载
0
收藏
0
当前安装
12
版本数
在 OpenClaw 中安装
/install openclaw-enterprise
功能描述
Use when user needs enterprise multi-agent collaboration system. Use when orchestrating 1 ChiefOfStaff + multiple specialized AI agents. Use when automating...
安全使用建议
This skill appears to be a coherent multi-agent orchestration design, but there are mismatches you should resolve before installing or providing credentials: 1) The package asks for ANTHROPIC_API_KEY but the included Python scripts do not use it — do not supply unrelated API keys unless you need them. 2) The repo lists server dependencies (FastAPI, langgraph) and npm metadata but has no install script — confirm how to deploy and which dependencies are actually required. 3) README examples show external API usage (api.openclaw.ai); if you will connect to such endpoints, review their privacy/security policies and consider using an isolated/test environment first. 4) The release notes indicate past security flags for hardcoded keys; scan the codebase for embedded secrets and network calls before use. Practical next steps: run a quick grep for 'API_KEY'/'Bearer' in the code, run the skill in an isolated container or VM, ask the publisher ([email protected] or the repository owner) to explain why ANTHROPIC_API_KEY is required and to provide an explicit install/deployment guide, and only provide credentials with least privilege (or use temporary/test keys).
功能分析
Type: OpenClaw Skill Name: openclaw-enterprise Version: 1.2.3 The skill bundle provides a comprehensive framework and simulation for an enterprise multi-agent orchestration system. The core logic in scripts like chief_of_staff.py, industrial_agents.py, and workflow_engine.py focuses on keyword-based task routing and workflow management using mock data, with no evidence of dangerous execution (eval/exec), unauthorized network access, or data exfiltration. While the documentation mentions requirements for API keys and external tools like curl, the implementation remains purely logical and aligned with its stated purpose of automating business workflows.
能力标签
cryptocan-make-purchasesrequires-oauth-tokenrequires-sensitive-credentials
能力评估
Purpose & Capability
The skill claims live LLM-driven multi-agent orchestration and shows examples calling external APIs (api.openclaw.ai, OpenAI). The included Python scripts, however, are mostly local simulators/mocks that do not actually call OpenAI or Anthropic. package.json lists server dependencies (FastAPI, uvicorn, langgraph) and Python >=3.10, but there is no install spec. These mismatches (claimed runtime vs shipped code and packaging) are unexplained and reduce confidence that the files reflect the documented production behavior.
Instruction Scope
SKILL.md and README provide normal usage instructions, triggers, and examples. They reference external APIs and webhook usage, but the runtime Python scripts operate locally and simulate agent responses. The runtime instructions do not ask the agent to read unrelated system files or exfiltrate data.
Install Mechanism
No install spec is provided (lowest direct install risk). However, package.json declares Python requirements and Node packaging metadata, and SKILL.md mentions installing via a ClawHub CLI — the mismatch between packaging metadata and lack of an install recipe is odd and could lead to users manually installing dependencies incorrectly or pulling additional packages from registries later.
Credentials
The registry metadata and SKILL.md require OPENAI_API_KEY and ANTHROPIC_API_KEY (with OPENAI_API_KEY as primary). The shipped code only reads OPENAI_API_KEY (default provider 'openai') and contains no usage of ANTHROPIC_API_KEY; the optional OPENCLAW_API_KEY is only referenced in docs. Requiring an Anthropic key while code does not reference it is an unexplained credential ask and therefore disproportionate.
Persistence & Privilege
Skill does not request always:true and uses default autonomous invocation settings; it does not modify other skills or system-wide configs. There is no evidence in the files that it attempts to persistently alter agent configuration or escalate privileges.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install openclaw-enterprise
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /openclaw-enterprise 触发
  4. 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
v1.2.3
Fix placeholder API keys in README
v1.2.2
Fixed install instructions: removed npm, use CLI only
v1.2.1
**Summary:** Major upgrade streamlining architecture, adding multi-agent extensibility, and improving documentation and setup. - Overhauled documentation with new files: quick start guides, security, pricing, publishing, and workflow examples. - Simplified and expanded pricing plans; added free, basic, team, enterprise, and custom tiers. - Switched default execution model to 16 specialized agents coordinated by a central "Chief of Staff" agent. - Enhanced setup instructions (requirements, environment, install scripts, examples) for faster onboarding. - Removed outdated templates, scripts, and references for leaner repo and easier maintenance. - RESTful API specifications and architecture diagrams added for easier integration.
v1.0.8
重新发布,等待安全审核
v1.0.7
修复安全问题:添加完整凭证声明、第三方域名显式声明
v1.0.6
openclaw-enterprise v1.0.6 - Added new script: scripts/industrial_agents.py - Updated package.json with latest changes - Enhanced support for industrial agent workflows through new script addition
v1.0.5
- Added comprehensive SKILL.md detailing system features, architecture, deployment, APIs, pricing, and agent roles. - Now includes clear multi-agent team structure, key collaboration workflows, and use-case triggers in both Chinese and English. - Provides precise installation and deployment instructions for SaaS and private options. - Outlines verification checklist and common rationalizations for reliable operation. - Updated metadata with API/environment/binary requirements and detailed pricing tiers.
v1.0.4
**Changelog for openclaw-enterprise v1.0.4** - Introduced a progressive loading structure in SKILL.md to optimize resource and instruction delivery. - Enhanced SKILL.md with improved, English-first descriptions and explicit trigger guidance for system use cases. - Added new "Common Rationalizations" and "Verification" sections in documentation to clarify system scope and expected outcomes. - Updated versioning and metadata to 1.0.4; adjusted author and tags display for consistency. - Added new release note file: 推广效果/重发布日志_20260424.md.
v1.0.3
License update: Unified to MIT-0. No functional changes.
v1.0.2
- Added homepage and license information to documentation - Expanded metadata to include API requirements, environment variables, and authentication details - Listed supported external APIs (OpenAI, Anthropic, Enterprise APIs) and their purposes - Specified required environment variables (OPENAI_API_KEY, ANTHROPIC_API_KEY) and binaries (python3, pip, curl) in metadata - No functional changes to the system described, but documentation is now more detailed and explicit for setup and API usage
v1.0.0
**OpenClaw Enterprise v1.0.0 - 全新简化与专注企业AI团队版本** - 精简定位为“1幕僚长+20专业Agent”企业运营AI系统,突出团队替代与协作能力 - 彻底重构&精简文档,移除大量说明、原快速入门、API/架构/细节和示例,简介聚焦应用价值与团队架构 - 全面优化定价体系,主打Basic/Professional/Enterprise三档,明晰并行数与功能覆盖 - 技术栈与要求简明化:仅需python3、pip、httpx等常规组件,去除复杂安装指令 - 明确卸载了旧API接口、文件组织和快速上手脚本,仅提供SaaS和私有化部署指导 - 关键词路由与职业Agent清单高度可读,便于企业用户理解与选型
v1.2.0
openclaw-enterprise 1.2.0 - 发布企业级多Agent协作系统,基于 LangChain + LangGraph 构建 - 提供16种专业执行Agent,支持复杂任务智能分解与并行处理 - 新增 Designer Agent,集成 Coze CLI 支持图片生成与在线交付 - 完善企业级特性:记忆系统、权限控制、成本追踪、审计日志 - 提供多档定价方案(免费至定制),满足不同企业需求 - 丰富快速上手指南与详细 API 文档
元数据
Slug openclaw-enterprise
版本 1.2.3
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 12
常见问题

Openclaw Enterprise 是什么?

Use when user needs enterprise multi-agent collaboration system. Use when orchestrating 1 ChiefOfStaff + multiple specialized AI agents. Use when automating... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 233 次。

如何安装 Openclaw Enterprise?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install openclaw-enterprise」即可一键安装,无需额外配置。

Openclaw Enterprise 是免费的吗?

是的,Openclaw Enterprise 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Openclaw Enterprise 支持哪些平台?

Openclaw Enterprise 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 Openclaw Enterprise?

由 WangM-A3(@wangm-a3)开发并维护,当前版本 v1.2.3。

💬 留言讨论