How does this skill help reduce LLM token costs?

The integrated Prompt Optimizer tool analyzes your prompts for redundancy and provides suggestions to consolidate instructions, potentially saving significant token usage while maintaining output quality.

Does it support agent workflow visualization?

Absolutely. The Agent Orchestrator can parse your agent configurations and generate ASCII or Mermaid diagrams to help you visualize and debug execution logic.

Can I use this to troubleshoot my RAG system?

Yes, the RAG Evaluator specifically measures context relevance and answer faithfulness, helping you determine if your system is suffering from poor retrieval or model hallucination.

Will this help with getting consistent JSON responses?

Yes, it includes a dedicated Structured Output Design workflow that combines schema definition with format enforcement patterns to ensure reliable, parseable responses.

Senior Prompt Engineer

Name: Senior Prompt Engineer
Author: oabdelmaksoud

byoabdelmaksoud

0•

데이터 과학 및 ML

Optimizes LLM performance through advanced prompt engineering, RAG evaluation frameworks, and agentic system orchestration.

The Senior Prompt Engineer skill provides a comprehensive toolkit for developers building sophisticated AI applications. It offers automated tools to analyze prompts for token efficiency, evaluate Retrieval-Augmented Generation (RAG) quality through metrics like faithfulness and relevance, and design complex agentic workflows. By implementing industry-standard patterns such as Chain-of-Thought, few-shot learning, and structured JSON output design, this skill helps bridge the gap between simple LLM queries and production-grade agentic systems with predictable, high-quality results.

주요 기능

01Few-shot example management and edge-case design workflows

02Structured output design with JSON schema validation

03RAG evaluation metrics for context relevance and answer groundedness

04Automated prompt optimization for token efficiency and clarity

050 GitHub stars

06Agent orchestration and visualization using ReAct and Plan-Execute patterns

사용 사례

01Reducing API costs by identifying and removing redundant prompt tokens

02Visualizing and validating complex multi-tool agent architectures

03Debugging RAG pipelines by measuring retrieval precision and generation faithfulness

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add oabdelmaksoud/agention senior-prompt-engineer

For use in Claude.ai and ChatGPT

주요 기능

01Few-shot example management and edge-case design workflows

02Structured output design with JSON schema validation

03RAG evaluation metrics for context relevance and answer groundedness

04Automated prompt optimization for token efficiency and clarity

050 GitHub stars

06Agent orchestration and visualization using ReAct and Plan-Execute patterns

사용 사례

01Reducing API costs by identifying and removing redundant prompt tokens

02Visualizing and validating complex multi-tool agent architectures

03Debugging RAG pipelines by measuring retrieval precision and generation faithfulness

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add oabdelmaksoud/agention senior-prompt-engineer

For use in Claude.ai and ChatGPT