Design and optimize prompts for LLM output quality and reliability.
npx skills add https://github.com/Estom/aiflex --skill prompt-engineer-estom
Please help me install this Agent Skill. Skill: prompt-engineer Source: https://github.com/Estom/aiflex/tree/main/skills-repo/Jeffallan-skills/prompt-engineer Command: npx skills add https://github.com/Estom/aiflex --skill prompt-engineer-estom
Designing and tuning prompts to produce reliable, cost-efficient, and validated outputs from large language models while reducing hallucinations, format errors, and model-specific inconsistencies.
Generate a prompt that returns a validated JSON summary with three prioritized action items and confidence scores for the following document: {document}
High-intent search queries and answers about installing and using this skill.
Design LLM prompts for structured outputs by defining schema constraints for JSON mode and function calling, applying validation examples, and implementing retry-with-correction flows to ensure reliable, format-compliant model responses.
Evaluate prompt quality and detect regressions using LLM-as-judge methods, automated test suites, A/B testing, and CI-driven regression checks to validate prompt variations and maintain production reliability.
Reduce LLM hallucinations using few-shot and chain-of-thought prompting by selecting practical templates that guide reasoning steps, applying ReAct or tree-of-thought patterns to improve output accuracy and consistency.
Optimize prompt token efficiency by applying an iterative optimization loop and token reduction techniques, using diagnostic frameworks to maintain output accuracy while lowering LLM API costs.
Yes, you can use Pydantic and Zod schema validation for LLM function calling outputs by designing structured schemas that enforce JSON format constraints and enable automated correction flows for invalid responses.
ContextJet-aiWrite evaluations before changing LLM prompts or models.
ollieb89Evaluate LLM outputs with automated metrics and LLM-as-Judge workflows.
JaiminVadadoriyaTest AI features including LLM prompts, tool calls, and nondeterministic outputs.
faionfaionGuide LLM integration, RAG pipelines, agent development, and model evaluation.
