AgentsMédio

Prompt Engineering Patterns

porneolabhq·neolabhq· v1.0.0 · atualizado em 2026-04-10

Score

Utilize this skill for writing commands, hooks, skills for Agents, or prompts for sub-agents or any other LLM interaction. This includes optimizing prompts, enhancing LLM outputs, and designing production-ready prompt templates.

prompt-engineeringllm-optimizationprompt-templatesfew-shot-learningchain-of-thoughtrag-integration

Linguagens

TypeScriptJavaScriptPythonJavaC#

0Stars

0Forks

0Usos

Fork

Documento do Skill

SKILL.mdcustomaize-agent-prompt-engineering/workflow

1. Define the desired LLM behavior and output format.

2. Select appropriate prompt engineering techniques (e.g., few-shot, chain-of-thought).

3. Craft an initial prompt based on the chosen techniques.

4. Test the prompt with diverse inputs and edge cases.

5. Evaluate the LLM's performance based on accuracy, consistency, and token usage.

6. Refine the prompt based on the evaluation results.

7. Implement version control for prompt management.

Telemetria de Agentes

Execuções

total

Taxa de Sucesso

últimos 30d

Latência Média

0.0s

p50

Alucinação

0.0%

detecção

Tokens Entrada

avg 0/exec

Tokens Saída

avg 0/exec

Uso por Plataforma

Skills Relacionados

Similar aSteve Jobs Perspective

60%

Similar aAmazon Working Backwards

60%

Similar aRoundtable Discussion

60%

Árvore do Skill

Prompt Engineering Patterns

customaize-agent-prompt-engineering

Fases Cognitivas6

1.SENSE

2.CONTEXTUALIZE

3.HYPOTHESIZE

4.EVALUATE

5.REFLECT

6.ACT

Triggers8

optimize a prompt for better LLM outputimprove the performance of a language modeldesign a prompt template for consistent resultsuse few-shot learning to guide LLM behaviorimplement chain-of-thought prompting for complex reasoningintegrate prompt engineering with a RAG systemwrite a system promptrefactor a prompt for token efficiency

Avaliar este Skill

Score Breakdown

⭐Avaliação Humana0%

🤖Sucesso de Agentes0%

🕐Atualidade100%

🔗Saúde de Dependências100%

🕸️Centralidade no Grafo0%

🛡️Segurança50%

CompositeScore = α·Humano + β·Agente + γ·Recência + δ·Deps + ε·Centralidade + ζ·Segurança

Instalação

$ synaptic mcp download customaize-agent-prompt-engineering

$ synaptic skills detail customaize-agent-prompt-engineering

$ synaptic skills live customaize-agent-prompt-engineering

Dependências

pdfplumber

Links

GitHub Repository