PPT Design MCP
Template-driven PowerPoint design MCP server for Windows, enabling slide creation and manipulation via COM automation.
README
PPT Design MCP
Template-driven PowerPoint design MCP server for Windows.
This project wraps Microsoft PowerPoint COM automation behind a Python adapter
and exposes a small MCP tool surface. MCP tools validate parameters, call the
adapter, and return a uniform result shape. COM calls stay in
ppt_design_mcp.adapters.
Requirements
- Windows desktop
- Microsoft PowerPoint installed
- Python 3.11+
pywin32- Official MCP Python SDK
Install
uv sync
If you are not using uv, install the project dependencies from
pyproject.toml into a Python 3.11+ environment.
Run the MCP Server
uv run ppt-design-mcp
Equivalent module form:
uv run python -m ppt_design_mcp.server
The server uses stdio transport through FastMCP's default run() behavior.
Exposed Tools
Every tool returns:
ok: booleantool: tool namemessage: human-readable status or errordata: structured payloaderror_code: error class name, ornullon success
Tools:
powerpoint_statusopen_template_as_newinspect_slidesduplicate_slideset_named_shape_textfill_named_shape_with_imageexport_slide_pngsave_and_close
Safety Notes
- Source templates are opened read-only and must not be overwritten.
- Generated files should be saved under
output/. - Shapes are addressed by stable shape names, not shape indexes.
- PowerPoint COM logic is not implemented in MCP tool functions.
Template Registry
Template page metadata is stored in local JSON manifest files. A manifest points to a PowerPoint template file and lists reusable template pages:
{
"template_file": "../output/presentations/com-smoke-test.pptx",
"templates": [
{
"template_id": "com-smoke-title",
"source_slide": 1,
"purpose": "Smoke-test title slide for adapter and MCP validation",
"style": "PowerPoint default title layout",
"slots": [
{
"shape_name": "Title 1",
"type": "text",
"max_chars": 80
}
]
}
]
}
Files under templates/ are ignored by Git in this repository so existing
designer templates and their manifests stay local unless they are reviewed and
published intentionally.
Use ppt_design_mcp.templates.load_template_registry() to load a manifest.
The registry exposes:
list_templates()get_template(template_id)check_consistency(inspector)
check_consistency() uses an object with list_slides() and list_shapes() to
verify that the manifest matches the real template. It reports:
- missing template files
- missing named shapes
- duplicate
template_idvalues - illegal
source_slidepage numbers
The registry does not perform page planning and does not call any large model.
Local Validation
py -3.12 -m pytest
py -3.12 scripts/build_demo.py
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.