senior-prompt-engineer
Master LLM Prompt Engineering
Auch verfügbar von: alirezarezvani
Creating effective prompts for large language models requires deep expertise in patterns, frameworks, and optimization techniques. This skill provides production-ready prompt engineering strategies for Claude, GPT-4, and other LLMs, including structured outputs, chain-of-thought reasoning, and agentic system design.
Die Skill-ZIP herunterladen
In Claude hochladen
Gehe zu Einstellungen → Fähigkeiten → Skills → Skill hochladen
Einschalten und loslegen
Teste es
Verwendung von "senior-prompt-engineer". Create a prompt that helps developers write clean, documented Python code with type hints
Erwartetes Ergebnis:
- Define clear role and expertise level for the AI
- Specify output structure and format requirements
- Include example code with inline comments
- Add validation criteria for successful outputs
- Provide guidelines for handling edge cases
Verwendung von "senior-prompt-engineer". Design a few-shot prompt for sentiment analysis of customer reviews
Erwartetes Ergebnis:
- Provide 3-5 labeled examples showing positive, negative, and neutral reviews
- Include brief explanations for each example classification
- Add format instructions for consistent response structure
- Specify handling for ambiguous cases
Verwendung von "senior-prompt-engineer". Create a chain-of-thought prompt for mathematical problem solving
Erwartetes Ergebnis:
- Break down the problem into explicit steps
- Show intermediate calculations and reasoning
- Verify each step before proceeding to the next
- Final answer with supporting work shown
Sicherheitsaudit
SicherDocumentation-focused skill containing reference guides and template Python scripts for prompt optimization, RAG evaluation, and agent orchestration. All scripts are skeleton implementations with standard file I/O only. No network calls, credential access, or shell command execution detected. All 68 static findings are false positives caused by scanner misinterpretation of documentation keywords and markdown code fence syntax.
Risikofaktoren
⚡ Enthält Skripte (3)
📁 Dateisystemzugriff (3)
Qualitätsbewertung
Was du bauen kannst
Build Production AI Systems
Design robust prompting pipelines for production AI applications with evaluation frameworks and monitoring strategies.
Define AI Product Requirements
Create clear prompt specifications and success metrics for AI-powered features and user experiences.
Optimize LLM Performance
Apply systematic evaluation techniques to improve model outputs and reduce costs through prompt optimization.
Probiere diese Prompts
Solve this problem step by step. First, identify the key components. Second, analyze each component. Third, reason through the relationships. Finally, derive the conclusion with supporting evidence.
Complete the following tasks based on these examples: Example 1: [input] -> [output]. Example 2: [input] -> [output]. Now complete: [new input] -> ?
Provide your response as a valid JSON object with these exact fields: { "summary": "brief summary", "confidence": 0.0-1.0, "reasoning": "explanation" }. Do not include any other text.You are an expert [role] with [years] years of experience. You always: 1) Start with clear understanding, 2) Provide structured responses, 3) Include practical examples, 4) Ask clarifying questions when needed.
Bewährte Verfahren
- Start with clear role definition and context setting for the AI model
- Use specific output formats like JSON or structured lists to reduce variability
- Iterate systematically by testing prompts and measuring results against metrics
- Include error handling and fallback instructions for edge cases
Vermeiden
- Avoid vague instructions that leave interpretation to the model
- Do not overload prompts with too many constraints simultaneously
- Avoid assuming the AI has context you have not explicitly provided
- Do not use prompts that try to circumvent safety guidelines