Featured

Deploy OpenClaw in 60 seconds — 20% off logoDeploy OpenClaw in 60 seconds — 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free
Your own AI agent, running 24/7 with QwikClaw logoYour own AI agent, running 24/7 with QwikClaw

QwikClaw sets up and runs an always-on OpenClaw agent for you. One click, no config files, no server setup.

Deploy now
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data — no proxies, no parsers, no maintenance.

Start building free
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it — secured from day one.

Get it set up for you
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free
Reach 47,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here

Works with

Claude CodeClaude DesktopCursorVS CodeClineCodex CLIOpenClaw+ any MCP client

Install to Claude Code

This server doesn't publish a one-line install command. Follow the setup in the source repository.

Summary

Enables AI assistants to create, modify, and manage PowerPoint presentations programmatically using python-pptx.

README.md

PowerPoint Automation MCP Server

A Model Context Protocol (MCP) server that provides comprehensive PowerPoint automation capabilities using python-pptx. This server enables AI assistants to create, modify, and manage PowerPoint presentations programmatically.

Demo

https://github.com/user-attachments/assets/3daf3bef-4d75-4639-a891-0e64b80b4807

Features

  • Presentation Management: Create, open, save, and close presentations
  • Slide Operations: Add, delete, and manage slides with various layouts
  • Content Creation: Add text boxes, images, tables, and charts
  • Text Formatting: Update text with font styling (name, size, bold, italic)
  • Shape Manipulation: Move, resize, and modify shape properties
  • Table Operations: Create tables and update cell content
  • Chart Creation: Add various chart types (column, line, pie, bar) with data
  • Group Management: Ungroup shapes for easier manipulation
  • Layout Support: Work with different slide layouts and placeholders

Recommended Prompts

Translation Scenario

For translating presentations, process in batches to avoid context overflow and ensure data safety:

Translate this presentation [absolute path of ppt], processing 10 slides at a time and saving after each batch of 10 slides.

This approach:

  • Prevents token limit issues on large presentations
  • Saves progress incrementally to avoid data loss
  • Allows for review between batches

Installation

git clone https://github.com/buckylee2019/powerpoint-automation-mcp.git
cd powerpoint-automation-mcp
uv sync

Running the Server

uv run --directory /path/to/powerpoint-automation-mcp mcp-ppt-server

API Reference

The server provides 25+ tools for comprehensive PowerPoint automation. All operations work with a single active presentation to simplify the API.

Presentation Management

initialize_powerpoint() -> bool

Initialize the PowerPoint automation system.

create_presentation(template: str = None) -> Dict

Create a new presentation, optionally from a template.

  • template: Optional path to template file

open_presentation(file_path: str) -> Dict

Open an existing PowerPoint file.

  • file_path: Full path to .pptx file

get_presentation() -> Dict

Get information about the currently active presentation.

save_presentation(path: str = None) -> Dict

Save the active presentation.

  • path: Optional save path (uses current path if omitted)

close_presentation() -> Dict

Close the active presentation.

Slide Management

get_slides() -> List[Dict]

Get list of all slides with metadata.

add_slide(layout_index: int = 1) -> Dict

Add a new slide with specified layout.

  • layout_index: Layout type (0=Title, 1=Title+Content, etc.)

delete_slide(slide_index: int) -> Dict

Delete a slide from the presentation.

  • slide_index: 0-based slide index

get_slide_layouts() -> Dict

Get all available slide layouts with their properties.

set_slide_title(slide_index: int, title: str) -> Dict

Set the title of a slide.

Content Inspection

get_slide_shapes(slide_index: int) -> Dict

ALWAYS RUN FIRST! Get all shapes in a slide with properties and IDs.

  • Returns shape types, positions, sizes, and text content
  • Identifies grouped shapes that may need ungrouping

get_slide_text(slide_index: int) -> Dict

Get all text content in a slide organized by shape.

  • Detects grouped shapes that may need ungrouping first

Text and Shape Manipulation

update_text(slide_index: int, shape_index: int, text: str, font_name: str = None, font_size: int = None, bold: bool = None, italic: bool = None, preserve_existing: bool = True) -> Dict

Update text content with optional formatting.

  • Supports font styling while preserving existing formatting
  • preserve_existing: Keep original formatting for unspecified properties

update_shape_by_id(slide_index: int, shape_id: str, text: str = None, left: float = None, top: float = None, width: float = None, height: float = None) -> Dict

Update shape properties by ID.

  • Positions and sizes in inches

Content Creation

add_textbox(slide_index: int, text: str, left: float = 1, top: float = 1, width: float = 4, height: float = 2) -> Dict

Add a text box to a slide.

add_image(slide_index: int, image_path: str, left: float = 1, top: float = 1, width: float = None, height: float = None) -> Dict

Add an image to a slide.

  • Maintains aspect ratio if only width or height specified

add_table(slide_index: int, rows: int, cols: int, left: float = 1, top: float = 1, width: float = 8, height: float = 4) -> Dict

Add a table to a slide.

add_chart(slide_index: int, chart_type: str, categories: List[str], series_names: List[str], series_values: List[List[float]], left: float = 1, top: float = 1, width: float = 8, height: float = 4, has_legend: bool = True) -> Dict

Add a chart with multiple data series.

  • chart_type: 'COLUMN', 'LINE', 'PIE', 'BAR'
  • categories: X-axis labels
  • series_names: Legend labels
  • series_values: Data values for each series

Table Operations

update_table_cell(slide_index: int, shape_index: int, row: int, col: int, text: str) -> Dict

Update text in a specific table cell.

get_table_content(slide_index: int, shape_index: int) -> Dict

Retrieve all content from a table.

Advanced Operations

ungroup_shapes(slide_index: int) -> Dict

Ungroup all text-containing groups in a slide.

  • Only processes simple groups (no nested groups)
  • Preserves shape positions and properties
  • Skips groups without text content

Usage Workflow

  1. Initialize: Call initialize_powerpoint()
  2. Open/Create: Use open_presentation() or create_presentation()
  3. Inspect: Always run get_slide_shapes() first to understand slide structure
  4. Ungroup if needed: Use ungroup_shapes() if grouped shapes are detected
  5. Modify: Use appropriate tools to add/modify content
  6. Save: Call save_presentation() to persist changes

Error Handling

All functions return dictionaries with either success data or error information:

// Success response
{
  "success": true,
  "message": "Operation completed",
  "data": {...}
}

// Error response
{
  "error": "Description of what went wrong"
}

MCP Configuration

Add to your MCP settings:

{
  "mcpServers": {
    "powerpoint-automation": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/path/to/powerpoint-automation-mcp",
        "mcp-ppt-server"
      ]
    }
  }
}

Dependencies

  • python-pptx: PowerPoint file manipulation
  • mcp: Model Context Protocol framework
  • fastmcp: Simplified MCP server creation

License

MIT License - see LICENSE file for details.

See related servers & alternatives →

Related MCP servers

Browse all →

Related guides

Hand-picked reading to help you choose and use AI & ML servers.