← 返回 Skills 市场
Prompt Debugger
作者
charlie-morrison
· GitHub ↗
· v1.0.0
· MIT-0
48
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install prompt-debugger
功能描述
Debug prompts that produce unexpected AI outputs — diagnose failure modes, identify ambiguity and conflicting instructions, test variations, compare model re...
安全使用建议
This skill appears coherent and appropriate for debugging prompts. Before using it, avoid pasting secrets, private keys, or proprietary data into prompts you want analyzed — the agent will need to send prompt text (and any example outputs) to the model(s) it uses, which could transmit that data. If you need to debug sensitive prompts, redact or syntheticize private data, or run prompt tests against a local/private model. If you prefer the agent not to test variations autonomously, restrict its ability to call external models or run the skill only interactively.
功能分析
Type: OpenClaw Skill
Name: prompt-debugger
Version: 1.0.0
The 'prompt-debugger' skill is a purely instructional Markdown-based tool designed to help users analyze and improve AI prompts. It contains no executable code, network requests, or file system access, and its instructions are strictly focused on prompt engineering diagnostics and optimization (SKILL.md).
能力评估
Purpose & Capability
The name/description (prompt debugging, diagnosis, variations, rewrites) matches the SKILL.md content. The skill requires no binaries, env vars, or installs — which is proportionate for an instruction-only prompt-debugging helper.
Instruction Scope
The runtime instructions are narrowly focused on analyzing prompts, classifying failure modes, proposing fixes, generating variations, and producing rewrites. They do not instruct the agent to read unrelated files, access environment variables, or transmit data to external endpoints beyond the normal model invocations required to test prompts. Note: applying the skill (especially 'compare' and 'test variations') will usually cause the agent to send prompt content and candidate outputs to whichever model or API the agent uses — so sensitive content could be transmitted as part of normal operation.
Install Mechanism
No install spec or code files; instruction-only skills are lowest-risk because nothing is written to disk or downloaded.
Credentials
The skill declares no environment variables, credentials, or config paths. That is proportional to its purpose.
Persistence & Privilege
always:false and no special persistence or system-wide config changes. The skill can be invoked by the agent (normal behavior) but does not request elevated presence or modify other skills.
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-debugger - 安装完成后，直接呼叫该 Skill 的名称或使用
/prompt-debugger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 prompt-debugger skill.
- Diagnose and fix prompt failures, including ambiguity, conflicting instructions, or missing context.
- Generate and compare prompt variations to test and improve prompt quality.
- Rewrite prompts with best practices (clarity, structure, format, examples).
- Score prompts across multiple quality dimensions and suggest top improvements.
- Detect and report common prompt anti-patterns.
- Includes a library of prompt patterns and anti-patterns for reference.
元数据
常见问题
Prompt Debugger 是什么？
Debug prompts that produce unexpected AI outputs — diagnose failure modes, identify ambiguity and conflicting instructions, test variations, compare model re... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 48 次。
如何安装 Prompt Debugger？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install prompt-debugger」即可一键安装，无需额外配置。
Prompt Debugger 是免费的吗？
是的，Prompt Debugger 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Prompt Debugger 支持哪些平台？
Prompt Debugger 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 Debugger？
由 charlie-morrison（@charlie-morrison）开发并维护，当前版本 v1.0.0。
推荐 Skills