Managing Documentation Pages

AI Docs automates the creation of comprehensive documentation from your GitHub repositories, but it also provides robust tools for managing and refining the generated content. This ensures that your documentation remains accurate, well-organized, and perfectly tailored to your project's needs. You can easily view, edit, reorder, and group documentation pages directly within the AI Docs platform.

Viewing Documentation Pages

Once AI Docs has processed your repository and generated documentation, you can access it through the "Docs" tab on your project's dashboard. This tab presents a structured view of all generated pages.

By default, pages are ordered based on the AI's initial assessment of logical flow and importance. You can select any page from the list to view its content rendered in a clean, readable format.

Editing Page Content

AI Docs empowers you to take full control over your documentation. While the AI generates initial content, you can refine, expand, or correct any page as needed. This is particularly useful for adding specific details, clarifying complex concepts, or incorporating information that might not be directly present in the codebase.

When you edit a page, you are directly modifying its Markdown content. These changes are saved instantly and reflected in your live documentation.

API Endpoint: /api/docs/update

The /api/docs/update endpoint facilitates the real-time editing of documentation page content.

• Purpose: To allow authenticated users to update the Markdown content of a specific generated documentation page.
• Method: POST
• Inputs:
  • id (string): The unique identifier of the documentation page to be updated.
  • content (string): The new Markdown content for the page.
• Process:
  1. The request is authenticated to ensure it comes from a logged-in user.
  2. Authorization checks verify that the specified documentation page exists and belongs to a project owned by the authenticated user.
  3. The content and updatedAt fields for the documentation page are updated in the database.

// Example of a conceptual client-side update request
async function updateDocumentationPage(docId: string, newContent: string) {
  try {
    const response = await fetch('/api/docs/update', {
      method: 'POST',
      headers: {
        'Content-Type': 'application/json',
      },
      body: JSON.stringify({ id: docId, content: newContent }),
    });

    if (!response.ok) {
      const errorData = await response.json();
      throw new Error(errorData.error || 'Failed to update documentation');
    }

    console.log('Documentation page updated successfully!');
    // Optionally, re-fetch the page to display updated content
  } catch (error) {
    console.error('Error updating documentation page:', error);
  }
}

// Usage example (assuming you have the docId and new content)
// updateDocumentationPage('doc_abc123', '# My Updated Title\n\nThis is the new content of my page.');

Reordering and Grouping Pages

To enhance the navigability and logical flow of your documentation, AI Docs allows you to reorder pages and organize them into custom groups. This is particularly useful for creating a hierarchical structure that guides users through your project's features and concepts.

You can typically drag-and-drop pages within the "Docs" tab to change their display order. You can also assign pages to specific groups, which might appear as sections or categories in the documentation sidebar.

API Endpoint: /api/docs/reorder

The /api/docs/reorder endpoint handles the bulk updating of documentation page order and grouping.

• Purpose: To allow authenticated users to define the display order and group assignments for multiple documentation pages within a project.
• Method: POST
• Inputs:
  • updates (array of objects): An array where each object contains:
    • id (string): The unique identifier of the documentation page.
    • order (number): The new numerical order for the page (e.g., 1, 2, 3...).
    • group (string, optional): The new group name for the page.
• Process:
  1. The request is authenticated.
  2. Authorization checks ensure all specified documentation pages belong to a single project owned by the authenticated user.
  3. A database transaction is used to atomically update the order and group fields for all specified pages, ensuring data consistency.

// Example of a conceptual client-side reorder request
async function reorderDocumentationPages(projectId: string, pageUpdates: Array<{ id: string; order: number; group?: string }>) {
  try {
    const response = await fetch('/api/docs/reorder', {
      method: 'POST',
      headers: {
        'Content-Type': 'application/json',
      },
      body: JSON.stringify({ updates: pageUpdates }),
    });

    if (!response.ok) {
      const errorData = await response.json();
      throw new Error(errorData.error || 'Failed to reorder documentation pages');
    }

    console.log('Documentation pages reordered successfully!');
    // Optionally, re-fetch the project docs to display the new order
  } catch (error) {
    console.error('Error reordering documentation pages:', error);
  }
}

// Usage example
// const updates = [
//   { id: 'doc_page_1', order: 1, group: 'Introduction' },
//   { id: 'doc_page_2', order: 2, group: 'Introduction' },
//   { id: 'doc_page_3', order: 1, group: 'Advanced Topics' },
// ];
// reorderDocumentationPages('project_xyz789', updates);

Interacting with Page Content

Beyond viewing and editing, AI Docs provides convenient tools for interacting with the content of a documentation page. These tools are typically found in a "Copy page" dropdown menu on the page itself.

• Copy page: This option copies the entire content of the current documentation page as Markdown text to your clipboard. This is particularly useful if you want to paste the content into an external editor or share it.
• View as Markdown: Opens the raw Markdown content of the page in a new browser tab. This allows you to inspect the underlying Markdown structure without any rendering.
• Open in ChatGPT / Claude / Perplexity: These options allow you to quickly send the current page's URL to popular Large Language Models (LLMs) like ChatGPT, Claude, or Perplexity. This pre-populates a prompt for the LLM to "read from [page URL]," enabling you to immediately ask questions about the page's content, summarize it, or get further explanations.

Best Practices for Managing Documentation Pages

• Regular Review: Even with AI generation, regularly review your documentation pages for accuracy, clarity, and completeness.
• Strategic Grouping: Use the grouping feature to create a logical and intuitive navigation structure. Think about how users will progress through your documentation.
• Leverage AI for Drafts, Refine Manually: Use AI Docs to quickly generate initial drafts, then apply your expertise to refine and perfect the content, ensuring it meets your specific standards and tone.
• Monitor Activity: Keep an eye on the Activity (Pipeline) tab after any regeneration to ensure all documentation pages are successfully updated.
• Utilize LLM Integration: For quick insights or to explore content from different angles, use the "Open in LLM" features to query the documentation with AI assistants.