Metadata-Version: 2.5
Name: mcpproducts-office
Version: 0.1.1
Summary: OfficeMCP — read, create and convert Office + PDF documents (local, pure Python)
Requires-Python: >=3.10
Requires-Dist: mcp>=1.2.0
Requires-Dist: openpyxl>=3.1
Requires-Dist: pypdf>=4
Requires-Dist: python-docx>=1.1
Requires-Dist: python-pptx>=0.6
Requires-Dist: reportlab>=4
Description-Content-Type: text/markdown

# OfficeMCP 📊📄📽️

**Read, create, and convert Office + PDF documents — local, free, pure Python.**

One server for the document chores every agent hits: extract PDF text, read/write Excel,
create Word docs with headings, outline PowerPoints, convert csv ↔ xlsx ↔ json.

## Tools

| Tool | Tier | What it does |
|---|---|---|
| `pdf_text` / `pdf_info` | free | Extract text / metadata from PDFs |
| `pdf_from_text` | free | Create a clean paginated PDF |
| `xlsx_read` / `xlsx_sheets` / `xlsx_write` | free | Excel read/write |
| `csv_head` / `csv_write` | free | CSV preview / write |
| `convert_tabular` | free | csv ↔ xlsx ↔ json conversion |
| `docx_text` / `docx_create` | free | Word read (headings+tables) / create (markdown-style) |
| `pptx_outline` | free | Slide-by-slide PowerPoint outline |
| `cloud_ocr_pdf` | cloud | OCR for scanned PDFs (needs `MPX_API_KEY`) |

## Install

**Claude Code:**
```bash
claude mcp add office -- uvx mcpproducts-office
```

**Claude Desktop** (`claude_desktop_config.json`):
```json
{
  "mcpServers": {
    "office": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/mkorayemai-web/office-mcp", "office-mcp"]
    }
  }
}
```

## Example prompts

- "Extract the text from report.pdf" → `pdf_text`
- "Create an Excel with columns Name/Score from this data" → `xlsx_write`
- "Turn these notes into a Word doc with headings" → `docx_create`
- "Convert data.csv to xlsx" → `convert_tabular`

## License

MIT

<!-- mcp-name: com.mcpproducts/office -->
