Dropbox Kb Auto

# Dropbox KB Auto

## What problem does this solve?

Your AI agent can't search your Dropbox. Documents, receipts, research papers, and notes sit in folders your agent doesn't know about — so you end up searching manually.

This skill bridges that gap: it syncs files from Dropbox, extracts text (PDF, Office, OCR for scans), and indexes everything into your agent's knowledge base. Changed files are detected via content hashing and automatically re-indexed. Unchanged files are never re-processed.

One-command installer configures folders, exclusions, file types, cron scheduling, and OpenClaw memory integration.

## What This Skill Does

```
    ┌───────────────────┐
    │  Dropbox Account  │
    └────────┬──────────┘
             │ Delta API (only changes)
             ▼
    ┌───────────────────┐
    │  Text Extraction  │
    │  PDF, Office, OCR │
    └────────┬──────────┘
             │ Markdown files
             ▼
    ┌───────────────────┐
    │  OpenClaw Memory  │
    │  Embed + Search   │
    └────────┬──────────┘
             │
             ▼
    ┌───────────────────┐
    │  Agent answers    │
    │  your questions   │
    └───────────────────┘
```

Supported formats: PDF (with OCR fallback), Word, Excel, PowerPoint, images (Tesseract OCR), plain text.

## Prerequisites

### System Dependencies
```bash
# Debian/Ubuntu
sudo apt-get install -y tesseract-ocr tesseract-ocr-eng poppler-utils

# macOS
brew install tesseract poppler
```

### Python Dependencies
```bash
pip3 install pypdf openpyxl python-pptx python-docx
```

Or run: `bash setup.sh`

### Dropbox App Setup

1. Create app at https://www.dropbox.com/developers/apps
2. Choose **Scoped access** → **Full Dropbox**
3. Enable: `files.metadata.read`, `files.content.read`
4. Generate refresh token via OAuth 2 flow
5. Add to `~/.openclaw/.env`:
   ```bash
   DROPBOX_FULL_APP_KEY=your_app_key
   DROPBOX_FULL_APP_SECRET=your_app_secret
   DROPBOX_FULL_REFRESH_TOKEN=your_refresh_token
   ```

## Installation

### Via...

# Dropbox KB Auto

## What problem does this solve?

You have years of documents, receipts, research papers, and notes in Dropbox — but your AI agent can't search any of them. Finding a specific file means opening Dropbox and digging through folders manually.

This skill connects Dropbox to your agent's knowledge base. It syncs your files, extracts text (including OCR for scans and photos), and indexes everything so you can ask natural-language questions like *"Find my Q1 2025 expense receipts"* and get answers instantly.

## How It Works

```
        ┌─────────────────────┐
        │   Dropbox (files)   │
        │  PDFs, Office, Imgs │
        └────────┬────────────┘
                 │  Delta API (only changes)
                 ▼
        ┌─────────────────────┐
        │   Text Extraction   │
        │                     │
        │  PDF → pypdf / OCR  │
        │  DOCX → python-docx │
        │  XLSX → openpyxl    │
        │  PPTX → python-pptx │
        │  IMG → Tesseract    │
        │  TXT → UTF-8 read   │
        └────────┬────────────┘
                 │  Saves as .md files
                 ▼
        ┌─────────────────────┐
        │  OpenClaw Memory    │
        │                     │
        │  Embeddings + Index │
        │  Semantic + keyword │
        │  hybrid search      │
        └────────┬────────────┘
                 │
                 ▼
        ┌─────────────────────┐
        │  Ask your agent:    │
        │  "Find my Q1 2025   │
        │   expense receipts" │
        └─────────────────────┘
```

## Features

- **Delta sync with content hashing** - Only processes new/changed files via Dropbox cursor API and content hashes (unchanged files are never re-downloaded or re-processed)
- **Multi-format extraction** - PDF, Word, Excel, PowerPoint, images, plain text
- **OCR** - Reads scanned documents and photos via Tesseract
- **Interactive installer** - Configures folders, exclusions, cron schedule in one command
- **Production-tested** - Handles...