Awesome Cursor Rules MDC — Auto-Generated .mdc Files

What it is

Awesome Cursor Rules MDC is an open-source repository that auto-generates .mdc (Markdown Cursor) rule files for 800+ libraries using Exa search and LLM processing. Each .mdc file teaches Cursor AI the conventions, best practices, and API patterns of a specific library so the editor produces higher-quality suggestions.

The project targets developers who use Cursor as their primary editor and want framework-aware code generation without manually writing custom rules for every dependency.

How it saves time or tokens

Without cursor rules, the AI assistant in Cursor relies on its training data alone, which may be outdated or generic. By loading a .mdc file that encodes the latest API surface and idioms for a library, you reduce the back-and-forth prompting needed to get correct code. Each rule file is compact enough to fit within Cursor's context budget, so it does not waste tokens on irrelevant information. The auto-generation pipeline keeps rules current as libraries evolve.

How to use

1. Browse rules at the repository: github.com/sanjeed5/awesome-cursor-rules-mdc
2. Find your framework or library in the rules-mdc/ directory
3. Copy the .mdc file into your project's .cursor/rules/ directory
4. Cursor auto-loads the rules on the next session

To generate a custom rule for a library not yet covered:

git clone https://github.com/sanjeed5/awesome-cursor-rules-mdc.git
cd awesome-cursor-rules-mdc
uv run src/generate_mdc_files.py --library 'fastapi'

Example

A generated .mdc file for FastAPI might look like this:

---
description: FastAPI best practices and patterns
globs: ['**/*.py']
alwaysApply: false
---

# FastAPI Rules
- Use Pydantic v2 models for request/response schemas
- Prefer async def for route handlers
- Use Depends() for dependency injection
- Return explicit status codes via status_code parameter
- Use APIRouter for route grouping

Drop this into .cursor/rules/fastapi.mdc and Cursor will apply these conventions whenever you work on Python files in that project.

Related on TokRepo

• Prompt Library — browse curated prompts and rules for various AI coding tools
• AI Tools for Coding — discover more developer productivity tools that integrate with AI editors

Common pitfalls

• Do not place .mdc files outside .cursor/rules/ or Cursor will not detect them
• Setting alwaysApply: true on many rules at once can bloat the context window and degrade suggestion quality
• Auto-generated rules may lag behind breaking API changes in fast-moving libraries; verify against official docs after generation