prompt-engineer
Skill0 starsChanged 7 months agoMITView on GitHub
Создание и оптимизация промптов для Claude, GPT и других LLM. Активируется на: промпт, prompt, системный промпт, инструкция для агента, create prompt, улучши промпт, напиши промпт, промпт для агента, промпт для задачи.
Files
Prompt Engineer — Мастер промптов
Универсальный инженер промптов для создания production-ready инструкций для любых языковых моделей.
Ключевой принцип: Ясность > Сложность. Неопределённость — главная причина провала промптов.
Параметры по модели
| Модель | Особенности | Оптимизация |
| Claude | XML теги, thinking | Структурированность |
| GPT | Function calling, JSON | Точные форматы |
| Gemini | Многомодальность | Контекст + медиа |
| o1/o3 | Reasoning models | Минимальные промпты |
Workflow
1. Анализ задачи
- Определение типа задачи
- Выбор подходящих техник
- Планирование структуры
2. Создание основы
- Роль и контекст
- Чёткие инструкции
- Примеры (few-shot)
3. Оптимизация
- Тестирование на крайних случаях
- Добавление ограничений
- Проверка форматирования
4. Валидация
- A/B тест вариантов
- Метрики качества
- Итерационные улучшения
Ключевые техники
1. Chain of Thought (CoT)
Think step by step:
1. Analyze the problem
2. Identify key factors
3. Apply reasoning
4. Provide conclusion2. Few-Shot Learning
Examples:
Input: [пример 1] → Output: [результат 1]
Input: [пример 2] → Output: [результат 2]
Now process: [новый input]3. Role Prompting
You are an expert [РОЛЬ] with [ОПЫТ].
Your task is to [ЗАДАЧА] using your expertise in [ОБЛАСТЬ].4. XML Structure (Claude)
<task>
[описание задачи]
</task>
<examples>
[примеры]
</examples>
<output_format>
[формат ответа]
</output_format>5. ReAct (Reasoning + Acting)
Thought: [что думаю]
Action: [что делаю]
Observation: [что получил]
... repeat until done
Final Answer: [итоговый ответ]6. Tree of Thoughts (ToT)
Generate 3 different approaches to solve this:
Approach 1: [решение 1]
Approach 2: [решение 2]
Approach 3: [решение 3]
Evaluate each approach and select the best one.Шаблоны промптов
Классификация
Classify the following [ОБЪЕКТ] into one of these categories: [КАТЕГОРИИ]
[ОБЪЕКТ]: [INPUT]
Category: [ВЫБОР ИЗ КАТЕГОРИЙ]
Confidence: [0-100%]
Reasoning: [ОБОСНОВАНИЕ]Анализ
Analyze the following [ДАННЫЕ] and provide insights on:
1. [АСПЕКТ 1]
2. [АСПЕКТ 2]
3. [АСПЕКТ 3]
Format as structured analysis with recommendations.Агент с инструментами
You have access to these tools:
- search(query): Search information
- calculate(expression): Perform calculations
- generate_image(prompt): Create images
Use tools when needed to complete the user's request.Оптимизация под модели
Claude (Anthropic):
- XML теги для структуры
- Thinking теги для рассуждений
- Чёткие role definitions
GPT (OpenAI):
- JSON форматы
- Function calling
- System/User/Assistant разделение
Gemini (Google):
- Мультимодальные промпты
- Длинные контексты
- Код + объяснения
Отладка промптов
| Симптом | Возможная причина | Решение |
| Неточные результаты | Нет примеров | Добавить few-shot |
| Разный формат ответов | Нет чёткой структуры | Задать output_format |
| Галлюцинации | Слишком творческая задача | Добавить ограничения |
| Медленная работа | Перегруженный промпт | Упростить инструкции |
| Не понимает контекст | Недостаточно контекста | Добавить background |
Метрики качества
| Метрика | Что измеряет |
| Точность | Правильность результатов |
| Консистентность | Стабильность ответов |
| Релевантность | Соответствие задаче |
| Полнота | Охват всех аспектов |
| Ясность | Понятность результата |
Частые ошибки
❌ Неясные инструкции — "сделай хорошо" вместо конкретики ❌ Отсутствие примеров — модель не понимает ожидания ❌ Перегруженность — слишком много задач в одном промпте ❌ Игнорирование формата — нет чёткой структуры ответа ❌ Амбивалентность — противоречивые требования ❌ Отсутствие контекста — модель не знает предметную область ❌ Нет обработки крайних случаев — промпт падает на исключениях
Other versions69 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 69
- Jeffallan/claude-skills★ 12ksame name
- elithrar/dotfiles★ 206same name
- chrispangg/deepagentsdk★ 128same name
- aakashg/pm-claude-skills★ 106same name
- thedesignproject/agent-skills★ 88same name
- OpenCoven/coven★ 75same name
- monaccode/astromesh★ 36same name
- Fandry96/k3-agentic-skills★ 27same name
- jshsakura/awesome-opencode-skills★ 27same name
- felvieira/claude-skills-fv★ 23same name
59 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
Used alongsideProjects with it also use idea-validator, debugger and 6 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
dipaulavsSkillidea-validator
Validates app ideas before building. Analyzes market saturation, competitors, real demand vs stated interest, solo builder feasibility (2-4 weeks timeline), and monetization potential. Use when user asks to validate an idea, evaluate a concept, check if idea is worth building, or before starting a new project.
Also in 3 projects that use this
JohnJBorenSkilldebugger
Systematic debugging workflows, root cause analysis, and troubleshooting patterns
Also in 3 projects that use this
VercelSkilltechnical-writer
Write, review, and edit developer documentation for SDKs, libraries, and frameworks, including sites such as ai-sdk.dev and chat-sdk.dev. Use for getting-started guides, tutorials, API references, conceptual explanations, integration guides, migration guides, troubleshooting, and code examples.
Also in 3 projects that use this
MicrosoftSkillapi-design
Guide for designing and documenting RESTful APIs. Use when asked to design an API, create endpoints, or document an API.
Also in 3 projects that use this
JetBrainsSkillwebapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
Also in 3 projects that use this
BlockSkillfrontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
Also in 6 projects that use this
In this repository35 other files it gives its agents.
- agent-architectSkill
- agent-memorySkill
- audience-lensSkill
- carouselSkill
- constructor-opsSkill
- content-repurposerSkill
- copywritingSkill
- customer-researchSkill
- editingSkill
- headlinesSkill
- heygenSkill
- learning-processorSkill
- memory-systemSkill
- nano-banana-pro-promptsSkill
- nano-bananaSkill
- openclaw-opsSkill
- personal-unpackingSkill
- reelsSkill
- selling-meaningsSkill
- seo-blog-writerSkill
- shared-learningsSkill
- skill-creatorSkill
- skill-securitySkill
- soul-masterySkill
- storytellingSkill
- swipefileSkill
- telegram-adminSkill
- telegramSkill
- text-postsSkill
- threadsSkill
- timelapse-creatorSkill
- tone-of-voiceSkill
- veo-videoSkill
- video-editorSkill
- youtubeSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

