How to Contribute Content

Cursor is an AI-powered code editor that can help you create content for the Prompt Library with built-in rules for formatting and structure.

Setting Up Cursor

1. Install Cursor

   Download and install Cursor from their official website.

2. Clone the repository

   git clone https://github.com/your-username/prompt_library.git

3. Open the project in Cursor

   Open the prompt_library folder in Cursor.

Using Cursor's AI Assistance

1. Create a new file in the appropriate directory based on your content type and discipline.
2. Leverage the prompt-library-requirements rule

   Cursor has built-in rules to help with content creation. You can ask the AI:

   "I want to create a new prompt for [your purpose]. Can you help me follow the prompt-library-requirements?"

3. Review and refine the generated content with the AI's help.
4. Submit your changes following the GitHub contribution workflow above.

If you have a content idea but don't want to create the files yourself, you can submit a content suggestion via GitHub Issues.

Submitting a Content Suggestion

1. Go to the Issues page of the repository.
2. Create a new issue with the title beginning with "Content Suggestion:".
3. Include the following information:
   • Content type (Prompt, Cursor Rule, Project Config, or Workflow State)
   • Discipline category (Development, Project Management, etc.)
   • Content title and description
   • Detailed content or a rough draft
   • Any reference materials or examples

What Happens Next

Maintainers will review your suggestion and may:

• Create the content based on your suggestion
• Request more information or clarification
• Suggest modifications to fit the library's format
• Invite you to contribute the content directly

You can now contribute external resources (guides, documentation, prompt lists, tools, etc.) to the library for any discipline.

How to Submit a Resource

1. Use the Resource Submission Issue Template on GitHub.
2. Fill out all required fields (title, description, URL, resource type, topics, discipline, etc.).
3. Submit the issue. A maintainer will review and add it to the appropriate discipline's resources/ folder.

What Qualifies as a Resource?

• External documentation or official guides
• Curated prompt lists
• Educational articles, videos, or tutorials
• Useful tools or reference sites

Maintainer Workflow

1. Review the submitted issue for completeness and relevance.
2. Create a Markdown file in the appropriate resources/ folder using the provided metadata.
3. Close the issue once the resource is added.

Skills live in the Lullabot/lullabot-skills repository and are automatically synced to this site when a pull request is merged. You don't need to touch this repository at all — just contribute upstream and your skill appears here within minutes.

Step-by-Step Process

1. Fork the upstream repository

   Fork Lullabot/lullabot-skills to your GitHub account and clone it locally.

2. Create a branch and a new folder

   git checkout -b add-my-skill
   mkdir my-skill-name

   The folder name becomes the skill's URL slug, so use lowercase kebab-case (e.g., code-review, drupal-recipe).

3. Add SKILL.md

   This is the Claude Code skill definition — what gets loaded into an AI assistant's context. It must include a YAML frontmatter block:

   ---
   name: my-skill-name
   description: |
     One or two sentences describing when to invoke this skill.
     Use when the user asks to do X or Y.
   allowed-tools:
     - Read
     - Edit
     - Bash
   ---
   
   # Skill Title
   
   Your skill instructions go here...

   name and description are required. allowed-tools is optional but recommended.

4. Add meta.yml

   This controls how the skill appears on this site:

   discipline: development
   title: "My Skill Display Title"
   date: "YYYY-MM-DD"
   tags:
     - relevant-tag
     - another-tag

   discipline must be one of: development, content-strategy, design, project-management, quality-assurance, sales-marketing.

5. Optionally add resource files

   Any scripts, config files, or reference documents you place in the folder (e.g., setup.sh, config.yml) will appear as downloads on the skill's page automatically.

6. Open a pull request

   Push your branch and open a PR against main in lullabot-skills. Once merged, the skill is synced here — no further action needed.

Changelog Entries

If your commit should appear in the skill's public changelog on this site, add a User-Facing-Change: trailer to your commit message:

Add initial my-skill-name skill

User-Facing-Change: Initial release with support for X and Y

Commits without this trailer (typo fixes, internal refactors) are excluded from the public changelog automatically.