mulmocast-vision
Enables LLMs to automatically generate presentation slides from natural language instructions using 80+ business-oriented templates, with auto-save and PDF export.
README
mulmocast-vision
Note: This repository has been archived. Development has moved to mulmocast-plus monorepo (
packages/mulmocast-vision).
mulmocast-vision is a tool that uses LLMs via MCP (Model Context Protocol) to automatically generate presentation slides, similar to PowerPoint.
With 80+ business-oriented slide templates, you can quickly create proposals, strategy decks, and other professional materials.
✨ Features
-
📑 80+ Rich Templates
Business-optimized designs for proposals, reports, and analysis documents -
⚡ Instant Generation
Create professional slides in seconds using LLMs via :contentReference[oaicite:1]{index=1} (MCP) -
⚙️ Simple Setup
Just add a small config to your MCP client (e.g. :contentReference[oaicite:2]{index=2}) to start -
💾 Auto Save & Easy Export
Outputs are automatically saved under~/Documents/mulmocast-vision/
Export instantly to PDF -
🎨 Fully Customizable
HTML-based templates allow fine-grained design tweaks or full redesigns
You can create documents like these in just a few minutes.
Sample AI Company Analysis Slide (PDF)
No Installation Needed
Runs directly with npx. No global install required.
Setup
Here is an example for Claude desktop. Add the following to your claude_desktop_config.json.
You can use similar settings for other MCPs.
// claude_desktop_config.json
"mulmocast-vision": {
"command": "npx",
"args": [
"mulmocast-vision@latest"
],
"transport": {
"stdio": true
}
}
That's all for the setup.
If the path to npx is not set, specify the full path.
If npx is not installed, please install it in advance.
Usage
- Launch an MCP-compatible client (e.g., Claude Desktop)
- Just give an instruction like "Compare corporate analysis of AI companies such as OpenAi Anthropic Replicate. About 20 slides.".
- The generated slides will be saved automatically under
~/Documents/mulmocast-vision/{date}
Currently available features include creating slides for each page, updating specified slides, generating a PDF of all slides, and generating a PDF for specified slides.
You can instruct these actions via prompts.
Output Examples
- Proposal (Problem → Solution → Value → Next Action)
- Business Model Canvas
- SWOT, PEST, and 3C Analysis
- Summary, Agenda, and Closing Slides
Logging
mulmocast-vision automatically logs all operations and errors to help with debugging and monitoring.
Log Location
Logs are saved in /tmp/mulmocast-vision-mcp/ with daily rotation:
- Format:
mcp_yyyymmdd.log(e.g.,mcp_20251206.log) - Each day creates a new log file
What is Logged
- MCP Server Operations: Server initialization, tool calls, and results
- File Operations: Template reads, HTML/PNG generation, PDF creation
- Errors: Detailed error messages with full stack traces
- Debug Information: Template rendering, directory operations, and more
Log Format
Logs are written in JSON Lines format for easy parsing:
{
"timestamp": "2025-01-15T10:30:45.123Z",
"level": "ERROR",
"message": "Template file not found",
"data": {
"errorMessage": "getHtml: file /path/to/template.html not exists.",
"errorName": "Error",
"stack": "Error: getHtml: file...\n at htmlPlugin.getHtml...",
"templateFilePath": "/path/to/template.html",
"functionName": "createAgendaSlide"
}
}
Custom Logger
You can replace the default logger with your own implementation (e.g., for telemetry):
import { setLogger, LoggerInterface } from 'mulmocast-vision/logger';
class CustomLogger implements LoggerInterface {
info(message: string, data?: unknown) {
// Your implementation
}
// ... other methods
}
setLogger(new CustomLogger());
For Developers
MulmoCast Vision is open source, so you can apply various designs by modifying the HTML. For adding styles, please refer to Style.ja.md.
Official Repository & Package
mulmocast-vision(日本語版)
mulmocast-vision は、MCP (Model Context Protocol) 経由で LLM を活用し、パワーポイントのようなスライド資料を自動生成できるツールです。 ビジネスに特化した 80種類のスライドテンプレート を備えており、企画書・提案資料・戦略資料などを簡単に作成できます。
特徴
- 📑 豊富なテンプレート: ビジネスに最適化された 80種類以上のスライドデザイン
- ⚡ 即時生成: LLM と MCP を通じて、最短数秒でプレゼン資料を出力
- 🖥️ シンプルな利用方法: MCP対応ツール(例: Claude Desktop)に設定するだけ
- 📂 自動保存: 生成された資料は
~/Documents/mulmocast-vision/以下に保存
インストール不要
追加のインストールは不要です。npx 経由で直接実行できます。
設定方法
MCP対応ツール(例: Claude Desktop)の設定ファイルに以下を追記してください。
"mulmocast-vision": {
"command": "npx",
"args": [
"mulmocast-vision"
],
"transport": {
"stdio": true
}
}
使い方
- MCP対応クライアント(Claude Desktop など)を起動
mulmocast-visionを呼び出してスライド生成を指示- 完成した資料が自動的に
~/Documents/mulmocast-vision/に保存されます
出力例
- 企画書(Problem → Solution → Value → Next Action)
- ビジネスモデルキャンバス
- SWOT分析、PEST分析、3C分析
- サマリースライド、アジェンダスライド、クロージングスライド
ログ機能
mulmocast-visionは、デバッグや監視を支援するため、すべての操作とエラーを自動的にログに記録します。
ログの保存場所
ログは /tmp/mulmocast-vision-mcp/ に日次ローテーションで保存されます:
- 形式:
mcp_yyyymmdd.log(例:mcp_20251206.log) - 日付が変わると新しいログファイルが作成されます
記録される内容
- MCPサーバー操作: サーバーの初期化、ツール呼び出し、実行結果
- ファイル操作: テンプレート読み込み、HTML/PNG生成、PDF作成
- エラー: 詳細なエラーメッセージとスタックトレース
- デバッグ情報: テンプレートレンダリング、ディレクトリ操作など
ログフォーマット
ログはJSON Lines形式で記録され、解析が容易です:
{
"timestamp": "2025-01-15T10:30:45.123Z",
"level": "ERROR",
"message": "Template file not found",
"data": {
"errorMessage": "getHtml: file /path/to/template.html not exists.",
"errorName": "Error",
"stack": "Error: getHtml: file...\n at htmlPlugin.getHtml...",
"templateFilePath": "/path/to/template.html",
"functionName": "createAgendaSlide"
}
}
カスタムロガー
デフォルトのロガーを独自の実装(テレメトリーなど)に置き換えることができます:
import { setLogger, LoggerInterface } from 'mulmocast-vision/logger';
class CustomLogger implements LoggerInterface {
info(message: string, data?: unknown) {
// 独自の実装
}
// ... その他のメソッド
}
setLogger(new CustomLogger());
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.