reindex-docs

alanben's avatarfrom alanben

Re-index all PDF and HTML documents, update index.html, and commit/push changes to the repository

⭐0stars🔀0forks📁View on GitHub🕐Updated Jan 8, 2026
[Files and Documents]

When & Why to Use This Skill

This Claude skill automates the maintenance of static document repositories by scanning, indexing, and organizing PDF and HTML files. It dynamically updates a central index.html with a structured document tree and synchronizes changes directly to a Git repository, ensuring documentation is always current, properly categorized, and searchable without manual intervention.

Use Cases

  • Automated Documentation Portals: Automatically update a central web index whenever new technical specifications or project reports are added to a repository, keeping the 'last updated' metadata current.
  • Client Deliverable Management: Organize multi-client project files into a strict hierarchy (Client > Project > Section) with reverse-chronological sorting to ensure stakeholders always see the latest documents first.
  • Static Site Maintenance: Streamline the workflow of updating static HTML file listings and document trees, including automatic Git commits and pushes with standardized attribution.
  • Repository Cleanup: Automatically identify and remove empty folders or outdated references within a document repository to maintain a clean and navigable file structure.
namereindex-docs
descriptionRe-index all PDF and HTML documents, update index.html, and commit/push changes to the repository

Reindex Documents Skill

You are a specialized agent for maintaining the document index in this static document repository.

Your Task

Re-index all PDF and HTML documents in the repository and update index.html, then commit and push the changes.

Process Steps

  1. Scan for all documents

    • Use Glob to find all PDF files: **/*.pdf
    • Use Glob to find all HTML files: **/*.html
    • Run these searches in parallel for efficiency

  2. Read current index.html

    • Read the existing index.html to understand the structure
    • Note the current documentTree JavaScript object structure

  3. Update index.html with new documents

    • Update the last-updated date to today's date (format: YYYY-MM-DD)
    • Update the documentTree object to include:
      • All newly found documents
      • Proper Client → Project → Section → Folder hierarchy
      • Documents sorted in reverse chronological order (newest first)
      • Clean, readable document titles
      • Relative paths from root

  4. Organizational Rules

    • Clients (top level): Turnbuckle, Nextgen, Loeries, MachinaXX, etc.
    • Projects (under clients): RentaStore, Adstream, LifeGauge, etc.
    • Sections (under projects): Meeting notes, Technical, Projects, Programs, etc.
    • Folders (under sections): Webstream, Phase 4, Modernization Program, etc.
    • Documents: Individual PDF and HTML files
    • Sort documents by date in reverse chronological order (most recent first)
    • Remove any empty folders that have no documents

  5. Commit and Push

    • Add all untracked files (new documents)
    • Commit index.html changes with descriptive message
    • Commit new document files with descriptive message
    • Push all commits to remote repository
    • Use proper commit message format with Claude Code attribution

Commit Message Format

Use this format for commits:

Update index with new documents - [Month Year]

Added/Updated documents:
- Client/Project: Document description
- Client/Project: Document description

Updated last modified date to YYYY-MM-DD

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

For new document files:

Add new client documents - [Month Year]

New documents added:
- Client/Project: Document description
- Client/Project: Document description

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Important Notes

  • Exclude these HTML files from the document index (they are system files):
    • index.html (the index itself)
    • repos.html (repository index)
    • access-denied.html (security page)
  • Maintain the existing JavaScript structure and styling
  • Preserve the auth.js script reference in the head section
  • Keep sections collapsed by default for clean navigation
  • Use relative paths for all document links
  • Follow the existing naming conventions for document titles

Expected Output

After completion, provide a summary of:

  • Number of new documents added
  • Number of clients/projects updated
  • Commit IDs created
  • Confirmation that changes were pushed successfully