An advanced prompt engineering skill that builds precise, high-performance prompts from scratch through structured dialogue and advanced reasoning frameworks.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install lyra-prompt-architect
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 lyra-prompt-architect 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 lyra-prompt-architect?
The lyra-prompt-architect skill is a highly sophisticated cognitive architect designed to move beyond basic prompt editing. Instead of making superficial tweaks, it deconstructs user goals to assemble complete, modular prompt architectures from the ground up. By utilizing strategic dialogue and functional formatting, it significantly reduces cognitive load while optimizing performance.
Built to elevate your interactions with advanced LLMs, this skill leverages professional prompt engineering methodologies to unlock adaptive intelligence. It dynamically adjusts its process based on the complexity of your task and your personal expertise level, ensuring every generated output is robust, reliable, and perfectly aligned with your core objectives.
lyra-prompt-architect Use Cases
- Designing robust, high-stakes prompts for fields like legal analysis, financial reporting, and complex data synthesis.
- Decomposing multifaceted enterprise tasks into structured, sequential subtasks.
- Mapping complex problem-solving routines to advanced cognitive frameworks like Tree-of-Thoughts or Graph-of-Thoughts.
- Building custom expert personas and contextual rulesets for target AI environments.
How lyra-prompt-architect Works
- Initialization Protocol: The skill launches with a structured greeting and pauses to collect vital parameters: the Target AI model and the preferred Optimization Level.
- Phase 1: Dialogue: A progressive, multi-turn dialogue engine captures deep system needs, intended audience tone, architectural constraints, and accuracy requirements.
- Phase 2: Blueprint Strategy: The agent selects the ideal reasoning framework (Chain-of-Thought, Tree-of-Thoughts, Graph-of-Thoughts, or Algorithm-of-Thoughts) that best matches the logic of the problem.
- Phase 3: Synthesis: Modular functional components (Role, Context, Task, Format, Examples, Constraints) are compiled into a cohesive, production-ready markdown layout.
- Phase 4: Refinement: The finalized prompt is presented alongside blueprint explanations, enhancements, and advanced verification loops like Chain-of-Verification (CoVe) if high fidelity is required.
lyra-prompt-architect Setup
Prerequisites
Ensure Openclaw Skills management tooling is configured on your development system.
Installation
Install the skill directly into your local AI agent ecosystem setup:
openclaw skills install lyra-prompt-architect
Activation
Initialize the cognitive architecture module within your prompt interface or workspace configurations. The script will automatically trigger the initial greeting protocol on its first turn.
lyra-prompt-architect Data Schema & Taxonomy
Input Configuration
Upon initialization, the system maps user parameters to a dedicated internal taxonomy:
| Parameter | Description | Valid Options |
|---|---|---|
target_ai |
The LLM execution engine running the generated prompt | ChatGPT-4, Claude 4, Gemini, etc. |
optimization_level |
Depth of dialogue and architecture execution path | Quick Boost, Deep Dive, Revolutionary |
Generated Architecture Schema
Outputs generated through this process are structured into precise modular sections inside a standard markdown payload:
- Role Definition: High-authority persona and domain expert assignment.
- Context Layer: Structured background documentation and guardrails.
- Task Decomposition: Granular, chronological steps detailing workflow logic.
- Format Specification: Strict templates or schema blueprints for target outputs.
- Meta-Cognitive Layers: Verification and self-correction modules appended to the final string.
lyra-prompt-architect Advanced Features
- Advanced Reasoning Integration: Selects and builds specialized cognitive networks including Chain-of-Thought (CoT), Tree-of-Thoughts (ToT), Graph-of-Thoughts (GoT), and Algorithm-of-Thoughts (AoT).
- Metacognitive Prompting (MP): Implements four-stage reflective loops (Understand -> Judge -> Assess -> Confirm) within prompts to handle sensitive logic.
- Chain-of-Verification (CoVe): Embeds self-contained fact-checking algorithms inside the prompt structure to minimize model hallucinations in critical tasks.
- Progressive Disclosure Dialogues: Minimizes developer fatigue by serving contextually aware, adaptive questions rather than large blocks of configuration requests.
SKILL.md
Loading
Related Openclaw Skills
D3.js Visualization
A powerful, low-level JavaScript library for binding arbitrary data to a Document Object Model and applying data-driven transformations.
Anthropic Claude API
A cost-conscious skill that empowers AI agents to securely call the Anthropic Claude API through specialized Model Context Protocol (MCP) server tools.
WeChat Mini Program CI (miniprogram-ci)
A headless compilation tool extracted from WeChat Developer Tools to automate Mini Program uploads, previews, and CI/CD operations without launching the GUI.
Review Feedback Schema
A structured, database-ready schema and verification workflow designed to track, audit, and systematically improve the accuracy of AI code reviews.
OpenAI Agent Skill
An operations guide and toolset for AI agents integrating OpenAI MCP services to generate text, compute embeddings, run reasoning tasks, produce images, and transcribe audio.
Tavily Web Search Skill
The Tavily Web Search Skill provides comprehensive instructional knowledge that teaches AI coding agents exactly when and how to leverage Tavily for live web searches, deep site crawls, content extraction, and mapping.
Featured*
