prompt-engineering

Name: prompt-engineering
Author: giuseppe-trisciuoglio

1.7kinstalls36forks

How to install prompt-engineering?

giuseppe-trisciuoglio/developer-kit/prompt-engineering

$npx -y skills add giuseppe-trisciuoglio/developer-kit --skill prompt-engineering

Installs into the current project.

Summary

Provides workflows to write, debug, and optimize prompts for LLMs, including few-shot example selection, chain-of-thought structuring, system prompt design, and template composition. Use when the user asks to write or improve a prompt, wants help with few-shot examples, chain-of-

Files · 1

View on GitHub

SKILL.md

1	# Prompt Engineering
2
3	## Overview
4
5	Use this skill to design prompt systems that are clear, testable, and reusable.
6	It covers prompt drafting, optimization, evaluation, and production-oriented
7	patterns for few-shot prompting, reasoning workflows, templates, and system
8	prompts.
9
10	Keep the main workflow in this file and load the targeted reference files only
11	for the pattern you are applying.
12
13	## When to Use
14
15	Use this skill when:
16
17	- A user asks to write, rewrite, or improve a prompt
18	- A prompt needs better structure, reliability, or output formatting
19	- Few-shot examples or reasoning scaffolds are needed
20	- A system prompt or reusable prompt template must be created
21	- An existing prompt needs measurable optimization and testing
22
23	Read the relevant files in `references/` when you need deeper guidance on a
24	specific pattern.
25
26	## Core Patterns
27
28	### 1. Few-Shot Learning
29
30	#### Example Selection Strategy
31	- Use `references/few-shot-patterns.md` for comprehensive selection frameworks
32	- Balance example count (3-5 optimal) with context window limitations
33	- Include edge cases and boundary conditions in example sets
34	- Prioritize diverse examples that cover problem space variations
35	- Order examples from simple to complex for progressive learning
36
37	#### Few-Shot Example (Sentiment Classification)
38	```
39	Classify the sentiment as Positive, Negative, or Neutral.
40
41	Text: "I love this product! It exceeded my expectations."
42	Sentiment: Positive
43	Reasoning: Enthusiastic language, positive adjectives, satisfaction
44
45	Text: "The app keeps crashing when I upload large files."
46	Sentiment: Negative
47	Reasoning: Complaint about functionality, frustration indicator
48
49	Text: "It arrived on time, as described."
50	Sentiment: Neutral
51	Reasoning: Factual statement, no strong emotion either way
52
53	Text: "{user_input}"
54	Sentiment:
55	Reasoning:
56	```
57
58	### 2. Chain-of-Thought Reasoning
59
60	#### Implementation Patterns
61	- Reference `references/cot-patterns.md` for detailed reasoning frameworks
62	- Use "Let's think step by step" for zero-shot CoT initiation
63	- Provide complete reasoning traces for few-shot CoT demonstrations
64	- Implement self-consistency by sampling multiple reasoning paths
65	- Include verification and validation steps in reasoning chains
66
67	#### CoT Template Structure
68	```
69	Let's approach this step-by-step:
70
71	Step 1: {break_down_the_problem}
72	Analysis: {detailed_reasoning}
73
74	Step 2: {identify_key_components}
75	Analysis: {component_analysis}
76
77	Step 3: {synthesize_solution}
78	Analysis: {solution_justification}
79
80	Final Answer: {conclusion_with_confidence}
81	```
82
83	### 3. Prompt Optimization
84
85	#### Optimization Process
86	- Use `references/optimization-frameworks.md` for comprehensive optimization strategies
87	- Measure baseline performance before optimization attempts
88	- Implement single-variable changes for accurate attribution
89	- Track metrics: accuracy, consistency, latency, token efficiency
90	- Use statistical significance testing for A/B validation
91	- Document optimization iterations and their impacts
92
93	Track these metrics: accuracy, consistency, token efficiency, robustness, safety. See `references/optimization-frameworks.md` for measurement utilities.
94
95	### 4. Template Systems
96
97	#### Template Design Principles
98	- Reference `references/template-systems.md` for modular template frameworks
99	- Use clear variable naming conventions (e.g., `{user_input}`, `{context}`)
100	- Implement conditional sections for different scenario handling
101	- Design role-based templates for specific use cases
102	- Create hierarchical template composition patterns
103
104	#### Template Structure Example
105	```
106	# System Context
107	You are a {role} with {expertise_level} expertise in {domain}.
108
109	# Task Context
110	{if background_information}
111	Background: {background_information}
112	{endif}
113
114	# Instructions
115	{task_instructions}
116
117	# Examples
118	{example_count}
119
120	# Output Format
121	{output_specification}
122
123	# Input
124	{user_query}
125	```
126
127	### 5. System Prompt Design
128
129	#### System Prompt Components
130	- Use `references/system-prompt-design.md` for detailed design guidelines
131	- Define clear role specification and expertise boundaries
132	- Establish output format requirements and structural constraints
133	- Include safety guidelines and content policy adherence
134	- Set context for background information and domain knowledge
135
136	#### System Prompt Framework
137	```
138	You are an expert {role} specializing in {domain} with {experience_level} of experience.
139
140	## Core Capabilities
141	- List specific capabilities and expertise areas
142	- Define scope of knowledge and limitations
143
144	## Behavioral Guidelines
145	- Specify interaction style and communi

People who liked this also viewed

6 picks

Type

prompt-engineering

1.7kinstalls36forks

View on GitHub

How to install prompt-engineering?

giuseppe-trisciuoglio/developer-kit/prompt-engineering

$npx -y skills add giuseppe-trisciuoglio/developer-kit --skill prompt-engineering

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Prompt Engineering
2
3	## Overview
4
5	Use this skill to design prompt systems that are clear, testable, and reusable.
6	It covers prompt drafting, optimization, evaluation, and production-oriented
7	patterns for few-shot prompting, reasoning workflows, templates, and system
8	prompts.
9
10	Keep the main workflow in this file and load the targeted reference files only
11	for the pattern you are applying.
12
13	## When to Use
14
15	Use this skill when:
16
17	- A user asks to write, rewrite, or improve a prompt
18	- A prompt needs better structure, reliability, or output formatting
19	- Few-shot examples or reasoning scaffolds are needed
20	- A system prompt or reusable prompt template must be created
21	- An existing prompt needs measurable optimization and testing
22
23	Read the relevant files in `references/` when you need deeper guidance on a
24	specific pattern.
25
26	## Core Patterns
27
28	### 1. Few-Shot Learning
29
30	#### Example Selection Strategy
31	- Use `references/few-shot-patterns.md` for comprehensive selection frameworks
32	- Balance example count (3-5 optimal) with context window limitations
33	- Include edge cases and boundary conditions in example sets
34	- Prioritize diverse examples that cover problem space variations
35	- Order examples from simple to complex for progressive learning
36
37	#### Few-Shot Example (Sentiment Classification)
38	```
39	Classify the sentiment as Positive, Negative, or Neutral.
40
41	Text: "I love this product! It exceeded my expectations."
42	Sentiment: Positive
43	Reasoning: Enthusiastic language, positive adjectives, satisfaction
44
45	Text: "The app keeps crashing when I upload large files."
46	Sentiment: Negative
47	Reasoning: Complaint about functionality, frustration indicator
48
49	Text: "It arrived on time, as described."
50	Sentiment: Neutral
51	Reasoning: Factual statement, no strong emotion either way
52
53	Text: "{user_input}"
54	Sentiment:
55	Reasoning:
56	```
57
58	### 2. Chain-of-Thought Reasoning
59
60	#### Implementation Patterns
61	- Reference `references/cot-patterns.md` for detailed reasoning frameworks
62	- Use "Let's think step by step" for zero-shot CoT initiation
63	- Provide complete reasoning traces for few-shot CoT demonstrations
64	- Implement self-consistency by sampling multiple reasoning paths
65	- Include verification and validation steps in reasoning chains
66
67	#### CoT Template Structure
68	```
69	Let's approach this step-by-step:
70
71	Step 1: {break_down_the_problem}
72	Analysis: {detailed_reasoning}
73
74	Step 2: {identify_key_components}
75	Analysis: {component_analysis}
76
77	Step 3: {synthesize_solution}
78	Analysis: {solution_justification}
79
80	Final Answer: {conclusion_with_confidence}
81	```
82
83	### 3. Prompt Optimization
84
85	#### Optimization Process
86	- Use `references/optimization-frameworks.md` for comprehensive optimization strategies
87	- Measure baseline performance before optimization attempts
88	- Implement single-variable changes for accurate attribution
89	- Track metrics: accuracy, consistency, latency, token efficiency
90	- Use statistical significance testing for A/B validation
91	- Document optimization iterations and their impacts
92
93	Track these metrics: accuracy, consistency, token efficiency, robustness, safety. See `references/optimization-frameworks.md` for measurement utilities.
94
95	### 4. Template Systems
96
97	#### Template Design Principles
98	- Reference `references/template-systems.md` for modular template frameworks
99	- Use clear variable naming conventions (e.g., `{user_input}`, `{context}`)
100	- Implement conditional sections for different scenario handling
101	- Design role-based templates for specific use cases
102	- Create hierarchical template composition patterns
103
104	#### Template Structure Example
105	```
106	# System Context
107	You are a {role} with {expertise_level} expertise in {domain}.
108
109	# Task Context
110	{if background_information}
111	Background: {background_information}
112	{endif}
113
114	# Instructions
115	{task_instructions}
116
117	# Examples
118	{example_count}
119
120	# Output Format
121	{output_specification}
122
123	# Input
124	{user_query}
125	```
126
127	### 5. System Prompt Design
128
129	#### System Prompt Components
130	- Use `references/system-prompt-design.md` for detailed design guidelines
131	- Define clear role specification and expertise boundaries
132	- Establish output format requirements and structural constraints
133	- Include safety guidelines and content policy adherence
134	- Set context for background information and domain knowledge
135
136	#### System Prompt Framework
137	```
138	You are an expert {role} specializing in {domain} with {experience_level} of experience.
139
140	## Core Capabilities
141	- List specific capabilities and expertise areas
142	- Define scope of knowledge and limitations
143
144	## Behavioral Guidelines
145	- Specify interaction style and communi

People who liked this also viewed

6 picks

Type