An advanced, multi-modal prompt engineering toolkit designed to optimize inputs for leading LLMs, image generation engines, and video AI models.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install prompt-engineering-plus
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install prompt-engineering-plus using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Prompt Engineering Plus?
Prompt Engineering Plus is an advanced framework and reference catalog designed to maximize the performance of AI models across text, image, and video domains. As a core contribution to the library of Openclaw Skills, this skill bridges the gap between basic queries and highly optimized, deterministic AI outputs. By providing structured guidelines for role prompting, chain-of-thought reasoning, constraint enforcement, and camera movement syntax, it empowers developers and prompt engineers to achieve production-grade consistency.
Whether you are writing system prompts for large language models like GPT-4 and Claude, directing visual generation engines like FLUX, or scripting cinematic camera movements for Veo, this repository equips you with model-specific rules of thumb. The inclusion of standardized evaluation checklists and iterative optimization pipelines ensures your generative workflows remain predictable and highly effective.
Prompt Engineering Plus Use Cases
- Code Auditing & Quality Control: Automatically structure code review requests to identify bugs, performance issues, and security vulnerabilities with specified output formats.
- Multi-Modal Content Generation: Generate precise image and video assets using highly structured style, composition, and temporal parameters for engines like FLUX and Veo.
- Data Analysis & Synthesis: Structure complex datasets into clean, schema-compliant JSON representations or narrative reports with high logical consistency.
- Prompt Performance Benchmarking: Establish a systematic evaluation pipeline to grade, audit, and iteratively improve prompt clarity, specificity, and output quality.
How Prompt Engineering Plus Works
- Determine Objective & Target Model: Identify whether the task requires text synthesis (Claude, GPT-4), visual design (FLUX), or cinematic video (Veo).
- Apply Architectural Structure: Build the base prompt using the standardized structural formula specific to the medium (e.g.,
[Subject] + [Style] + [Composition] + [Lighting] + [Technical]for images). - Inject Advanced Techniques: Layer in role prompting, constraint frameworks, and few-shot examples to constrain the output variance.
- Execute Model-Specific Optimization: Refine syntax based on the target model's strengths, such as utilizing 'let's think step by step' for GPT-4 or cinematic motion words for Veo.
- Evaluate & Iterate: Run outputs against the prompt quality checklist, score effectiveness across key metrics, and iteratively expand constraints and examples until the desired outcome is systematically achieved.
Prompt Engineering Plus Setup
To integrate Prompt Engineering Plus into your project, you can deploy it as part of your Openclaw Skills configuration.
Installation
Install the skill package using your preferred package manager or clone it directly into your local Openclaw directory:
git clone https://github.com/opencode/prompt-engineering-plus.git
cd prompt-engineering-plus
Configuration
No external API keys are strictly required for the templates, but you should reference these templates within your agent configuration to enforce standardized prompts. Update your LLM agent's system profile or utility scripts to point to the local reference schemas:
# Export the path to make templates globally accessible by your scripts
export OPENCLAW_PROMPTS_DIR="./prompt-engineering-plus/templates"
Prompt Engineering Plus Data Schema & Taxonomy
The Prompt Engineering Plus skill organizes its prompt architectures, evaluation metrics, and optimization matrices using standardized layouts. The layout taxonomy is detailed below:
| Component | File / Schema Purpose | Structure Format |
|---|---|---|
| LLM Schema | Structures system roles, Chain-of-Thought reasoning, and JSON constraints. | Markdown & Python Code Blocks |
| Image Schema | Organizes style keywords, compositional settings, and negative prompts. | Structured Text Blocks |
| Video Schema | Outlines tracking, panning, zoom, and temporal transitions. | Structural Formulas |
| Metrics Schema | Details effectiveness scores (Clarity, Specificity, Consistency, Quality). | Evaluation Markdown Tables |
| Templates | Ready-to-use patterns for Code Reviews, Content Writing, and Data Analysis. | Variable-interpolated Code Blocks |
Prompt Engineering Plus Advanced Features
- Model-Specific Optimization Layers: Custom formatting adapters designed to exploit the native behavioral quirks of Claude, GPT-4, Gemini, FLUX, and Veo.
- Iterative Optimization Workflow: A structured 5-step blueprint that guides prompts from basic queries to production-ready, highly constrained inputs.
- Negative Prompt Frameworks: Comprehensive style and quality exclusions optimized specifically for deep-learning image generators to eliminate artifacts and deformities.
- Structured Evaluation Metrics: A quantitative scoring rubric to rate visual and text generation performance, ensuring consistent output across high-throughput pipelines.
- Variable-Driven Templates: Parameterized blueprint configurations for software engineering, content generation, and statistical analysis tasks that can be easily automated within multi-agent setups.
SKILL.md
Loading
Related Openclaw Skills
Skill Creator Plus
An enhanced skill creation framework providing templates, automated quality gates, validation rules, and direct ClawHub publishing integration to build high-quality AI agent skills.
Boss Fight Planner Skill
A structured game design utility providing repeatable templates, checklists, and safe defaults for planning epic boss encounters.
Room Status Query
A developer tool that queries real-time and cached meeting room availability and recently released slots.
Plan Skill
The Plan Skill empowers AI agents to generate structured, step-by-step implementation blueprints and interactive sub-agent plan-deepeners before writing code.
Web Scrape Plus
An enhanced data extraction tool for AI agents featuring stealth browser automation, proxy routing, rate limiting, and robust structural data validation.
Data Analysis Plus
An enhanced data analysis kit providing comprehensive Python and R code templates, statistical methods, visualization guides, and automated reporting formats.
Featured*
