slidecraft-mcp

slidecraft-mcp

Enables agents to create and reconstruct editable PowerPoint presentations from source material, managing project resources and slide construction through natural language.

Category
访问服务器

README

Slidecraft

Slidecraft is a presentation creation framework that turns source material into a planned, visually designed, and fully editable PowerPoint deck.

It gives the creation process a clear structure from research and storyline development through slide design, editable reconstruction, and final assembly. A shared web workspace lets you manage the presentation, review progress, control the design system, and continue revisions across sessions.

What the framework adds

  • A research and briefing method that starts with the audience and intended decision
  • A storyboard with a recommended slide count and one conclusion-led message per slide
  • A consistent presentation design system covering typography, color, density, icons, connectors, and deck chrome
  • Shared materials, visual assets, reusable components, and visual references
  • Slide-level visual design with exact content and semantic relationships
  • Conversion of designed slides into editable PowerPoint content
  • A persistent project record for review, revision, and continuation
  • A central web app for presentation settings, resources, progress, and deliverables

How Slidecraft works

1. Build the presentation brief

Slidecraft guides the interpretation of your materials and any supporting research. The brief identifies the audience, the change in understanding the deck should create, the governing answer, the evidence available, and the assumptions that matter.

For collaborative work, you see the research synthesis and proposed brief before slide production begins.

2. Plan the argument

The framework compares possible narrative structures and develops a storyboard. The proposal includes the recommended number of slides, storyline phases, and a project-specific message, evidence allocation, visual job, and transition for every slide.

This makes the deck reviewable as an argument before time is spent designing pages.

3. Resolve the visual system

Slidecraft applies the selected communication style, information density, typography, colors, icon treatment, connector conventions, and page structure across the deck. Project assets and reusable visual resources remain attached to the presentation and available for later revisions.

4. Design each slide

Every slide begins with a communication contract that defines what the audience should understand and what the visual must make clear. Every information-bearing slide uses image generation for composition, including technical structures, tables, comparisons, and process diagrams. Covers and section transitions use consistent native layouts.

Each candidate is reviewed for message fidelity, evidence accuracy, hierarchy, grouping, relationships, legibility, and consistency with the deck.

5. Reconstruct editable PowerPoint content

Slidecraft identifies the meaningful text, shapes, tables, charts, images, icons, groups, and connectors in an accepted design. It measures their geometry and appearance, maps them to appropriate PowerPoint objects, fits typography, restores canonical assets, and rebuilds semantic connector systems.

OpenCV handles ordinary visual measurement. SAM can support irregular filled boundaries. The resulting slide remains editable in PowerPoint.

6. Assemble and review the deck

The framework assembles slides in storyboard order and checks package integrity, text fit, asset resolution, connector behavior, and repeated design roles. The final review covers the argument, evidence, visual rhythm, editability, and cross-slide coherence.

Shared workspace and web app

Each presentation has one project workspace containing its materials, assets, settings, working artifacts, and deliverables. The conversation and web app use the same project state.

The web app provides a central view for:

  • Presentation progress and latest outputs
  • Source material and project visuals
  • Selected icons, components, and visual inspiration
  • Communication style, density, typography, and color
  • Image-generation configuration
  • Editable slides and assembled decks

Open it with:

slidecraft console

Working with Slidecraft

You can stay involved throughout the process or delegate the complete presentation.

For collaborative work, ask to review the research synthesis and proposed brief first.

Create an executive presentation from these materials. Show me the research findings, recommended slide count, storyline, and message for every slide before designing the deck.

At any point, you can request the brief, storyboard, visual design, editable slide, or complete presentation.

Install

Install Python 3.10 or newer and the current Node.js LTS release. Then run the guided installer.

macOS and Linux

curl -fsSL https://raw.githubusercontent.com/henryhyw/slidecraft/v0.1.0-alpha.1/install.py | python3 -

Windows PowerShell

irm https://raw.githubusercontent.com/henryhyw/slidecraft/v0.1.0-alpha.1/install.py | py -3 -

See Installation for platform locations and troubleshooting.

For contributors

git clone https://github.com/henryhyw/slidecraft.git
cd slidecraft
python3 install.py --source .

Implementation details are documented in Agent integration, Presentation workflow, and Full-deck architecture.

License

Slidecraft is released under the MIT License. Bundled third-party resources retain their own notices.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选