← 返回 Skills 市场
prompt-optimizer-en
作者
ucsdzehualiu
· GitHub ↗
· v1.0.0
· MIT-0
118
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install prompt-optimizer-en
功能描述
Iterative prompt optimizer for complex tasks. Strictly implements ACON's two-stage iterative optimization + APE automatic prompt engineering. Only triggers w...
安全使用建议
This skill appears coherent and does only prompt transformation work, but remember: (1) it will see the full prompt you give it, so do not include secrets or sensitive data in prompts; (2) its validations (functional equivalence, effectiveness scoring) are subjective — review optimized prompts before using them in high-stakes contexts; and (3) because it rewrites prompts automatically when you request iterations, keep a copy of your original prompt if you need to revert or audit changes.
功能分析
Type: OpenClaw Skill
Name: prompt-optimizer-en
Version: 1.0.0
The skill bundle contains instructions for an AI agent to perform iterative prompt optimization based on academic methodologies (ACON and APE). The SKILL.md file defines a structured process for intent extraction, utility enhancement, and compression without any evidence of malicious intent, data exfiltration, or unauthorized command execution.
能力评估
Purpose & Capability
The name/description (prompt optimizer implementing ACON/APE) align with the SKILL.md: all operations are prompt parsing, candidate generation, scoring, compression, and iterative refinement. There are no extra credentials, binaries, or config paths requested that would be unrelated to prompt optimization.
Instruction Scope
Instructions stay within the stated purpose and explicitly forbid auto-triggering and extra analysis. One operational ambiguity: several validation steps require the agent to judge 'functional equivalence' and 'effectiveness' which are subjective and grant the agent discretion in deciding when a change is acceptable. This is expected for a prompt-optimization skill but means outputs should be reviewed by the user rather than assumed correct.
Install Mechanism
Instruction-only skill with no install spec and no code files; nothing is written to disk or downloaded by the skill itself.
Credentials
No environment variables, credentials, or config paths are requested. The skill does not request unrelated secrets or cloud credentials.
Persistence & Privilege
always is false and user-invocable only. The skill does not request persistent presence or modification of other skills or agent-wide settings.
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-optimizer-en - 安装完成后，直接呼叫该 Skill 的名称或使用
/prompt-optimizer-en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-optimizer skill with strict two-stage iterative optimization workflow.
- Implements ACON's (arXiv:2510.00615) input parsing, critical signal extraction, and compression methodology.
- Integrates APE (arXiv:2211.01910) automatic prompt engineering with multi-candidate generation and scoring.
- Supports multi-round optimization based on explicit user feedback, ensuring functional integrity at every step.
- Only activates when explicitly requested by the user and always preserves all critical signals and original task intent.
元数据
常见问题
prompt-optimizer-en 是什么？
Iterative prompt optimizer for complex tasks. Strictly implements ACON's two-stage iterative optimization + APE automatic prompt engineering. Only triggers w... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 118 次。
如何安装 prompt-optimizer-en？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install prompt-optimizer-en」即可一键安装，无需额外配置。
prompt-optimizer-en 是免费的吗？
是的，prompt-optimizer-en 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
prompt-optimizer-en 支持哪些平台？
prompt-optimizer-en 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-optimizer-en？
由 ucsdzehualiu（@ucsdzehualiu）开发并维护，当前版本 v1.0.0。
推荐 Skills