prompt-engineer
davila7/claude-code-templates
Design and optimize prompts for LLM applications with structured techniques and systematic evaluation.
What is prompt-engineer?
Expert guidance on crafting effective prompts for language models, covering system prompt architecture, context management, few-shot examples, and chain-of-thought reasoning. Use when building LLM-powered applications or debugging model behavior through prompt iteration.
- Design and optimize prompts for consistent, reliable LLM outputs
- Structure system prompts with role, context, instructions, constraints, and format specifications
- Manage context windows efficiently and curate relevant information
- Specify output formats explicitly to ensure parseable results
- Evaluate prompts systematically and measure impact of changes
- Create few-shot examples and chain-of-thought reasoning patterns
How to install prompt-engineer
npx skills add https://github.com/davila7/claude-code-templates --skill prompt-engineer- Understanding of LLM fundamentals
- Basic knowledge of tokenization
- Programming experience
How to use prompt-engineer
- 1.Define the model's role and context clearly in your system prompt
- 2.Structure instructions with explicit do's and don'ts rather than vague guidance
- 3.Include 2-5 diverse few-shot examples matching your expected input difficulty
- 4.Specify output format precisely (JSON, markdown, structured text, etc.)
- 5.Use chain-of-thought prompting by asking the model to reason step-by-step
- 6.Evaluate prompt changes systematically rather than iterating by intuition
- 7.Consider prompt injection risks when incorporating user input
Use cases
- Building reliable chatbot or assistant system prompts with clear role and constraints
- Designing few-shot examples to guide model behavior on classification or extraction tasks
- Implementing chain-of-thought prompting to debug model reasoning failures
- Optimizing context usage when working with large documents or knowledge bases
- Defending against prompt injection by structuring user input handling
- LLM application developers
- AI product managers
- Backend engineers integrating LLMs
- Anyone building prompt-based automation
prompt-engineer FAQ
A system prompt defines the model's role, constraints, and general behavior. Few-shot examples demonstrate desired output format and reasoning on specific tasks. Both are complementary—use system prompts for broad guidance and few-shot examples for task-specific behavior.
Include 2-5 diverse examples that cover the range of expected inputs and edge cases. More examples aren't always better; focus on quality and diversity rather than quantity.
Use chain-of-thought when you need to debug model reasoning, handle complex multi-step problems, or improve accuracy on tasks requiring logical steps. Ask the model to 'think step by step' and parse reasoning separately from the final answer.
Treat user input as untrusted data. Use clear delimiters between system instructions and user input, validate input format, and consider using structured prompts that separate instructions from user content.
Temperature depends on task: use lower values (0.0-0.3) for factual, deterministic tasks like extraction or classification; use higher values (0.7-1.0) for creative tasks. Test systematically rather than using defaults.
Full instructions (SKILL.md)
Source of truth, from davila7/claude-code-templates.
name: prompt-engineer description: "Expert in designing effective prompts for LLM-powered applications. Masters prompt structure, context management, output formatting, and prompt evaluation. Use when: prompt engineering, system prompt, few-shot, chain of thought, prompt design." source: vibeship-spawner-skills (Apache 2.0)
Prompt Engineer
Role: LLM Prompt Architect
I translate intent into instructions that LLMs actually follow. I know that prompts are programming - they need the same rigor as code. I iterate relentlessly because small changes have big effects. I evaluate systematically because intuition about prompt quality is often wrong.
Capabilities
- Prompt design and optimization
- System prompt architecture
- Context window management
- Output format specification
- Prompt testing and evaluation
- Few-shot example design
Requirements
- LLM fundamentals
- Understanding of tokenization
- Basic programming
Patterns
Structured System Prompt
Well-organized system prompt with clear sections
- Role: who the model is
- Context: relevant background
- Instructions: what to do
- Constraints: what NOT to do
- Output format: expected structure
- Examples: demonstration of correct behavior
Few-Shot Examples
Include examples of desired behavior
- Show 2-5 diverse examples
- Include edge cases in examples
- Match example difficulty to expected inputs
- Use consistent formatting across examples
- Include negative examples when helpful
Chain-of-Thought
Request step-by-step reasoning
- Ask model to think step by step
- Provide reasoning structure
- Request explicit intermediate steps
- Parse reasoning separately from answer
- Use for debugging model failures
Anti-Patterns
❌ Vague Instructions
❌ Kitchen Sink Prompt
❌ No Negative Instructions
⚠️ Sharp Edges
| Issue | Severity | Solution |
|---|---|---|
| Using imprecise language in prompts | high | Be explicit: |
| Expecting specific format without specifying it | high | Specify format explicitly: |
| Only saying what to do, not what to avoid | medium | Include explicit don'ts: |
| Changing prompts without measuring impact | medium | Systematic evaluation: |
| Including irrelevant context 'just in case' | medium | Curate context: |
| Biased or unrepresentative examples | medium | Diverse examples: |
| Using default temperature for all tasks | medium | Task-appropriate temperature: |
| Not considering prompt injection in user input | high | Defend against injection: |
Related Skills
Works well with: ai-agents-architect, rag-engineer, backend, product-manager
Related skills
More from davila7/claude-code-templates and the wider catalog.
pubmed-database
Direct REST API access to PubMed. Advanced Boolean/MeSH queries, E-utilities API, batch processing, citation management. For Python workflows, prefer biopython (Bio.Entrez). Use this for direct HTTP/REST work or custom API implementations.
pytorch-lightning
High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.
qdrant-vector-search
High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.
quality-documentation-manager
Senior Quality Documentation Manager for comprehensive documentation control and regulatory document review. Provides document management system design, change control, configuration management, and regulatory documentation oversight. Use for document control system implementation, regulatory document review, change management, and documentation compliance verification.
remotion
Programmatic video creation in React with frame-based animations, compositions, and media handling.
scientific-brainstorming
Research ideation partner. Generate hypotheses, explore interdisciplinary connections, challenge assumptions, develop methodologies, identify research gaps, for creative scientific problem-solving.