← 返回 Skills 市场
18630
总下载
92
收藏
218
当前安装
2
版本数
在 OpenClaw 中安装
/install prompt-engineering-expert
功能描述
Advanced expert in prompt engineering, custom instructions design, and prompt optimization for AI agents
安全使用建议
Reasonable to install as a prompt engineering reference. Review or revise the refusal-troubleshooting section before using it in safety-sensitive environments, and do not use fictional, academic, or professional framing to bypass legitimate model safety refusals.
功能分析
Type: OpenClaw Skill
Name: prompt-engineering-expert
Version: 1.0.0
The skill bundle is a comprehensive documentation set for an AI agent to act as an expert in prompt engineering. All files are markdown or JSON, containing no executable code. The content consistently focuses on best practices, techniques, and examples for crafting effective AI prompts. Crucially, `CLAUDE.md` and `README.md` explicitly state limitations, such as 'Does not execute code or run actual prompts (analysis only)' and 'Cannot access real-time data or external APIs', directly countering potential malicious activities. Furthermore, the skill actively addresses 'Jailbreak Vulnerabilities' in `CLAUDE.md` and `docs/BEST_PRACTICES.md`, indicating an intent to prevent, rather than perform, prompt injection. There is no evidence of data exfiltration, malicious execution, persistence, or any other high-risk behaviors.
能力评估
Purpose & Capability
The artifacts coherently match the stated purpose: prompt analysis, prompt generation, custom instruction design, best practices, examples, and troubleshooting.
Instruction Scope
The skill is instruction-only and explicitly states it does not execute code, run prompts, access real-time data, or call external APIs; one refusal-troubleshooting example uses fictional reframing around manipulation, which should be treated cautiously.
Install Mechanism
The bundle consists of markdown documentation and metadata; no package install scripts, binaries, declared dependencies, or required credentials are present.
Credentials
The skill does not request filesystem, network, account, API, credential, or broad local indexing authority beyond user-supplied prompt text.
Persistence & Privilege
No persistence mechanism, background worker, privilege escalation, autonomous mutation, or session/profile access is shown in the artifacts.
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 prompt-engineering-expert - 安装完成后，直接呼叫该 Skill 的名称或使用
/prompt-engineering-expert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 of the Prompt Engineering Expert skill.
- Provides in-depth guidance on prompt engineering, custom instruction design, and prompt optimization.
- Includes best practices, advanced techniques, evaluation methods, and context management strategies.
- Offers use cases such as prompt refinement, system prompt creation, and debugging.
- Added comprehensive documentation: GETTING_STARTED.md, INDEX.md, START_HERE.md, SUMMARY.md.
v0.1.0
- Initial release of the "Prompt Engineering Expert" skill.
- Provides comprehensive expertise in prompt engineering, instructions design, and prompt optimization.
- Supports advanced techniques such as chain-of-thought prompting, few-shot examples, and multimodal prompts.
- Offers best practices, anti-patterns identification, and prompt performance evaluation.
- Enables refining, debugging, and templating prompts for consistent AI agent behavior.
元数据
常见问题
Prompt Engineering Expert 是什么？
Advanced expert in prompt engineering, custom instructions design, and prompt optimization for AI agents. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 18630 次。
如何安装 Prompt Engineering Expert？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install prompt-engineering-expert」即可一键安装，无需额外配置。
Prompt Engineering Expert 是免费的吗？
是的，Prompt Engineering Expert 完全免费（开源免费），可自由下载、安装和使用。
Prompt Engineering Expert 支持哪些平台？
Prompt Engineering Expert 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 Prompt Engineering Expert？
由 TomsTools（@tomstools11）开发并维护，当前版本 v1.0.0。
推荐 Skills