← 返回 Skills 市场
AI Agent Runtime
作者
yesong-Hue
· GitHub ↗
· v1.2.0
· MIT-0
38
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install ai-agent-runtime
功能描述
A lightweight AI Agent runtime tool that provides an integrated web management interface, supports multiple LLM providers, and is fully compatible with OpenC...
使用说明 (SKILL.md)
AI Agent Runtime
A lightweight AI Agent runtime with web management interface and multi-LLM support
Features
- Lightweight design with minimal resource usage
- Built-in web-based management interface
- Multi-provider LLM support (OpenRouter, OpenAI, Anthropic, Ollama)
- SOUL.md configuration compatibility with OpenClaw
- Multi-agent simultaneous operation capability
- Optional local LLM support for offline operation
Installation
openclaw skills install ai-agent-runtime
After installation, access http://localhost:18953 to configure and use.
Resources
- More OpenClaw Skills: http://ai.qnitgroup.com
- ShadowAI API: https://referer.shadowai.xyz/r/1056448
安全使用建议
Review the installer source and release provenance before installing, avoid direct curl|bash or irm|iex execution if possible, and be prepared for the runtime to store API keys, conversation history, and agent memory under ~/.fastclaw.
功能分析
Type: OpenClaw Skill
Name: ai-agent-runtime
Version: 1.2.0
The skill bundle acts as an installer for an external tool ('FastClaw') by downloading and executing binaries from a third-party GitHub repository (fastclaw-ai/fastclaw) using high-risk patterns like 'curl | bash' in scripts/install.sh and recommending PowerShell 'iex' in README.md. It also includes financial referral links (shadowai.xyz) and manages sensitive configuration files (apikeys.json) locally. While these actions align with the stated purpose of a runtime manager, the use of unverified remote binaries and high-privilege installation methods presents a significant supply chain risk.
能力标签
能力评估
Purpose & Capability
The stated purpose, web UI, multi-provider LLM support, multi-agent operation, and local storage are coherent for an AI agent runtime.
Instruction Scope
The instructions are mostly user-directed, but the runtime supports persistent memories and multiple agents, so users should understand what information agents retain.
Install Mechanism
The README recommends curl|bash and PowerShell irm|iex remote installers, and the included script downloads an external release binary without checksum or signature verification.
Credentials
LLM API keys and local conversation data are expected for this kind of runtime, but they are sensitive and stored under the user's home directory.
Persistence & Privilege
The runtime stores configuration, API keys, agent files, conversation history, and MEMORY.md locally; this is disclosed and purpose-aligned but persistent.
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 ai-agent-runtime - 安装完成后，直接呼叫该 Skill 的名称或使用
/ai-agent-runtime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
v1.2.0
完全重写为英文，移除中文和敏感词，避免风险提示
元数据
常见问题
AI Agent Runtime 是什么？
A lightweight AI Agent runtime tool that provides an integrated web management interface, supports multiple LLM providers, and is fully compatible with OpenC... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 38 次。
如何安装 AI Agent Runtime？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install ai-agent-runtime」即可一键安装，无需额外配置。
AI Agent Runtime 是免费的吗？
是的，AI Agent Runtime 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
AI Agent Runtime 支持哪些平台？
AI Agent Runtime 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 AI Agent Runtime？
由 yesong-Hue（@yesong-hue）开发并维护，当前版本 v1.2.0。
推荐 Skills