AI SummaryHeuristic scoring (no AI key configured).
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to add the "cursor_rules_example — Cursor Rules" prompt rules to my project. Repository: https://github.com/Kryst4lskyxx/cursor_rules_example Please read the repo to find the rules/prompt file, then: 1. Download it to the correct location (.cursorrules, .windsurfrules, .github/prompts/, or project root — based on the file type) 2. If there's an existing rules file, merge the new rules in rather than overwriting 3. Confirm what was added
Description
A set of useful cursor rules
Basic Requirements
• Three-layer structure: basic/ (must apply), modules/ (apply as needed), workflow/ (apply as needed) • Use rule format in every file: Basic Requirements, Mandatory Actions, Forbidden Behaviors, Example Code • Prefer example-driven guidance: link to .mdc examples where helpful • Enforce consistent architecture, terminology, and quality gates across all outputs • Apply prompt engineering principles: clarity, directness, step-by-step reasoning (see Prompt Clarity, AI Reasoning)
Communication Best Practices
• Golden Rule: If a human colleague would be confused by your explanation, rephrase it. Be clear and direct. • Be Specific: Avoid vague statements like "ensure high performance" without measurable criteria • Ask When Unclear: When requirements are ambiguous, ask targeted clarifying questions immediately • Skip Preambles: Get straight to the point; avoid unnecessary introductions unless context is needed • Show Your Work: For complex tasks, explain reasoning step-by-step before conclusions
Mandatory Actions
• Identify scenario → select relevant rules from basic/, modules/, and workflow/ • Read example code/snippets → use as reference for style and patterns • Apply Mandatory and Forbidden items → do not bypass constraints • Validate output → typesafe, readable, testable, maintainable, secure, accessible • Resolve conflicts by priority → basic > modules > workflow; document exceptions • Record assumptions explicitly when context is missing; avoid fabricating APIs or data • For complex multi-step tasks: break down the problem, reason through each step, then provide solution • When handling user data/input: use clear separators (XML tags preferred) to distinguish from instructions • When uncertain about facts, APIs, or implementation details: admit uncertainty and suggest verification steps
Forbidden Behaviors
• Mixing roles or ignoring priority; contradicting rule layers • Vague guidance (e.g., “ensure high performance”) without measurable criteria • Introducing unapproved dependencies, APIs, or architectures • Omitting tests, types, or error handling in critical paths
Discussion
Health Signals
My Fox Den
Community Rating
Sign in to rate this booster
Works With
Any AI assistant that accepts custom rules or system prompts