Design, optimize, and evaluate prompts for LLM performance and safety.
npx skills add https://github.com/lamb92009/claude-skills --skill prompt-engineer-lamb92009
Please help me install this Agent Skill. Skill: prompt-engineer Source: https://github.com/lamb92009/claude-skills/tree/main/prompt-engineer Command: npx skills add https://github.com/lamb92009/claude-skills --skill prompt-engineer-lamb92009
Expert prompt engineer offering systematic design, optimization, and evaluation of prompts to maximize LLM performance, reliability, and safety, including advanced prompting techniques like chain-of-thought, few-shot, ReAct, and structured outputs.
Draft a role and task definition, then outline evaluation criteria and test prompts for edge cases to validate performance.
High-intent search queries and answers about installing and using this skill.
Build an LLM evaluation framework by establishing testing pipelines and guardrails that validate prompt behavior. This ensures repeatable, measurable, and safe outputs across various prompting scenarios.
Implement chain-of-thought and ReAct prompting by incorporating structured outputs and multi-step reasoning designs. These advanced techniques guide the LLM through logical steps, maximizing performance and ensuring measurable behavior.
Build an LLM evaluation framework by establishing testing pipelines and guardrails that validate prompt behavior. This ensures repeatable, measurable, and safe outputs across various prompting scenarios.
Yes, you can use structured outputs and guardrails to enforce repeatable, measurable prompt behavior. These governance frameworks ensure reliable and safe outputs across different LLM applications.
You need few-shot prompting when providing examples improves output accuracy, whereas zero-shot prompting relies on direct task definitions. Both methods are supported to maximize LLM performance based on scenario complexity.
VjayRamDefine DSPy signatures and modules with deterministic evaluation across splits.
v1bh0rImplement production readiness controls for monitoring, runbooks, and quality gates.
truchotIntegrate LLMs, RAG, and vector databases into web applications.
pferretti99Create or refine Claude skill directories with SKILL.md and evaluation loops.
