← Back to Skills Marketplace
102
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install prompt-expert
Description
Expert in prompt engineering, custom instructions, and prompt optimization for AI agents: clear structure, CoT/few-shot/XML/role patterns, system prompts, an...
Usage Guidance
This skill is documentation and templates only — safe in that it requests no credentials or installs nothing. Still: do not paste secrets, private data, or sensitive files into prompts when using it; test revised prompts in a safe environment before using for high‑stakes or production tasks; and review any generated system prompts or agent instructions for unintended escalation or data-leakage before deploying.
Capability Analysis
Type: OpenClaw Skill
Name: prompt-expert
Version: 1.0.0
The 'prompt-expert' skill bundle is a collection of documentation and templates focused on prompt engineering best practices. It contains no executable code, scripts, or instructions that attempt to exfiltrate data, achieve persistence, or perform unauthorized actions. All files (SKILL.md, docs/GUIDE.md, and examples/EXAMPLES.md) are strictly informational and aligned with the stated purpose of assisting in prompt optimization and agent instruction design.
Capability Assessment
Purpose & Capability
Name/description (prompt engineering, optimization, templates) match the SKILL.md, docs, and examples. There are no unrelated env vars, binaries, or install steps.
Instruction Scope
SKILL.md and docs contain only guidance, templates, and best practices for prompts and agent instructions. They do not instruct the agent to read system files, call external endpoints, exfiltrate data, or access credentials. Mentions of handling 'files/PDFs' or 'vision' are guidance for prompt formats, not code that executes file access.
Install Mechanism
No install spec and no code files — instruction-only, so nothing is downloaded or written to disk during install.
Credentials
No required environment variables, credentials, or config paths are declared or used. The skill does not request secrets or tokens.
Persistence & Privilege
Default invocation settings (always: false, model invocation allowed) are appropriate for a user-invocable guidance skill. The skill does not request persistent system-level presence or modify other skills.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install prompt-expert - After installation, invoke the skill by name or use
/prompt-expert - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of "prompt-expert" skill, providing expert guidance on prompt engineering and optimization.
- Offers best practices for clear, structured prompts, including advanced techniques like Chain-of-Thought, few-shot, XML, and role-based prompting.
- Covers custom instruction and system prompt design, iterative refinement, evaluation frameworks, and anti-pattern identification.
- Supports analysis and debugging for prompts involving multimodal inputs, embeddings, and agent tool-use.
- Comes with reference guides and curated examples for practical use and learning.
Metadata
Frequently Asked Questions
What is Prompt Expert?
Expert in prompt engineering, custom instructions, and prompt optimization for AI agents: clear structure, CoT/few-shot/XML/role patterns, system prompts, an... It is an AI Agent Skill for Claude Code / OpenClaw, with 102 downloads so far.
How do I install Prompt Expert?
Run "/install prompt-expert" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Prompt Expert free?
Yes, Prompt Expert is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Prompt Expert support?
Prompt Expert is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Prompt Expert?
It is built and maintained by mike47512 (@mike47512); the current version is v1.0.0.
More Skills