Feature Request: Bulk Documentation Import for Cursor AI

Feature Request: Bulk Documentation Import for Cursor AI

Overview

Currently, adding documentation URLs to Cursor AI requires manually adding them one at a time through the settings interface. For developers working on complex projects that require multiple documentation sources, this process is time-consuming and inefficient. We propose adding functionality for bulk documentation imports to streamline this process.

Use Case

When working on sophisticated projects (like our NBA DFS system), developers often need to reference multiple documentation sources:

  • Core language/framework docs (Python, FastAPI, etc.)
  • Data processing libraries (Pandas, NumPy, etc.)
  • Visualization tools (Matplotlib, Plotly, etc.)
  • Database documentation (MongoDB, PostgreSQL, etc.)
  • API documentation (Sports APIs, Weather APIs, etc.)

Manually adding these 50+ documentation sources one by one is time-consuming and prone to errors.

Proposed Solutions

Option 1: JSON/YAML Import

Allow users to import documentation URLs via a structured file:

docs:
  - name: "Pandas Documentation"
    url: "https://pandas.pydata.org/docs/"
    category: "Data Processing"
  
  - name: "FastAPI Documentation"
    url: "https://fastapi.tiangolo.com/"
    category: "Web Framework"

Option 2: Bulk URL Paste

Add a text area in settings where users can paste multiple URLs:

https://pandas.pydata.org/docs/
https://fastapi.tiangolo.com/
https://numpy.org/doc/stable/

Option 3: Documentation Presets

Provide preset documentation bundles for common development stacks:

  • Data Science Stack (Pandas, NumPy, Scikit-learn, etc.)
  • Web Development Stack (FastAPI, SQLAlchemy, etc.)
  • ML/AI Stack (PyTorch, TensorFlow, etc.)

Benefits

  1. Time Savings: Dramatically reduces time spent on documentation setup
  2. Improved Accuracy: Reduces potential for errors in URL entry
  3. Better Collaboration: Teams can share documentation configurations
  4. Standardization: Helps maintain consistent documentation access across projects
  5. Enhanced Onboarding: New team members can quickly set up their development environment

Technical Considerations

  1. URL Validation: Bulk import should validate URLs before adding
  2. Duplicate Detection: System should handle duplicate URLs gracefully
  3. Categories/Tags: Support for organizing documentation by category
  4. Export Functionality: Allow exporting current documentation settings
  5. Version Control: Enable documentation configurations to be version controlled

Implementation Suggestions

  1. Add an “Import” button next to the current “Add new doc” button
  2. Support multiple import formats (JSON, YAML, plain text)
  3. Provide a preview of URLs to be imported before confirming
  4. Add bulk export functionality for sharing configurations
  5. Include error handling for invalid URLs or format issues

Impact

This feature would significantly improve the development experience for:

  • Teams working on large projects
  • Developers managing multiple projects
  • Open source contributors
  • New developers setting up their environment
  • Organizations standardizing their development tools

Requested Priority

Medium-High: This feature would provide immediate value to users working on complex projects while not disrupting current functionality.

Community Benefits

  • Encourages sharing of documentation configurations
  • Helps standardize development environments
  • Reduces friction in project setup
  • Enables better team collaboration

Additional Considerations

Documentation Categories

Consider implementing a category system:

categories:
  - name: "Data Processing"
    docs:
      - "https://pandas.pydata.org/docs/"
      - "https://numpy.org/doc/stable/"
  
  - name: "Web Framework"
    docs:
      - "https://fastapi.tiangolo.com/"
      - "https://www.djangoproject.com/docs/"

Version Management

Allow specifying documentation versions:

docs:
  - url: "https://pandas.pydata.org/docs/"
    versions:
      - "2.0"
      - "1.5"
      - "1.4"

Tags and Search

Enable tagging and searching of documentation:

docs:
  - url: "https://pandas.pydata.org/docs/"
    tags: ["python", "data", "analytics"]

Discussion Points

  1. Preferred import format(s)?
  2. Should categories be predefined or user-defined?
  3. How to handle version management?
  4. Integration with existing project configuration files?
  5. Potential for documentation sharing platform?

Would love to hear the community’s thoughts on this feature request and any additional suggestions for implementation.


Note: The examples provided are based on a real-world NBA DFS (Daily Fantasy Sports) project requiring extensive documentation resources.

I like it…

One question, can you now use a bunch of MCPs for this purpose? It appears exactly what it could do…

Like have the agent slurp up docs for context based on rules?