Content
# Dynamo MCP 🚀
<p align="center">
<img src="assets/logo.png" alt="Dynamo MCP Logo" width="200"/>
</p>
<p align="center">
<b>The Self-Extending MCP Server</b><br/>
One MCP server to build them all ⚡
</p>
---
## 🌟 Why Dynamo MCP?
Most MCP servers are **static**: you code tools once, deploy, and that’s it.
**Dynamo MCP is different** — it’s the **world’s first self-extending MCP server**.
- 🧠 **LLM-powered tool generation** → Missing a tool? Dynamo asks an LLM to create one.
- 📂 **Persistent registry** → Every generated tool is saved & reusable forever.
- 🔌 **Plug-and-play for AIs** → Any AI (Claude, GPT, local models) can request tools.
- 🛡️ **Sandbox-ready** → Tools run in safe environments.
- 🌍 **Global utility hub** → From time zones to PDF generation — one server, infinite growth.
---
## 📊 System Architecture
```mermaid
graph TD
A[AI Client] -->|Request Tool| B[Dynamo MCP Server]
B -->|Check| C[Tool Registry]
C -->|Exists| D[Return Tool]
C -->|Missing| E[LLM Generator]
E -->|Generate| F[Dynamic Loader]
F -->|Register| C
D -->|Execute| G[Result Back to Client]
```
---
## 🔄 Workflow Example
```mermaid
sequenceDiagram
participant AI as AI Client
participant D as Dynamo MCP
participant R as Tool Registry
participant L as LLM Generator
AI->>D: Request "time_in_timezone"
D->>R: Check tool
R-->>D: Tool Found
D-->>AI: Return result ✅
AI->>D: Request "text_to_pdf"
D->>R: Check tool
R-->>D: Not Found ❌
D->>L: Generate tool
L-->>D: Returns code
D->>R: Save + Register tool
D-->>AI: Return result ✅
```
---
## 🚀 Quick Start
### 1️⃣ Clone the repo
```bash
git clone https://github.com/yourname/dynamo-mcp.git
cd dynamo-mcp
```
### 2️⃣ Install dependencies
```bash
pip install -r requirements.txt
```
### 3️⃣ Run the server
```bash
python main.py
```
### 4️⃣ Connect to your AI
- Claude Desktop → add `dynamo-mcp` in config
- OpenAI client → use MCP stdio transport
- Or call the HTTP endpoint directly
---
## ⚙️ Example Usage
**AI asks:**
> "What’s the time in Tokyo?"
✔️ Dynamo finds `time_in_timezone` → returns instantly.
**AI asks:**
> "Convert this text into a PDF."
❌ Tool not found →
⚡ Dynamo generates `text_to_pdf` with an LLM, saves it, and returns the PDF.
✔️ Next time → tool is already available.
---
## 📈 Roadmap
- [x] Core MCP server
- [x] Persistent registry of tools
- [ ] Sandbox execution (Docker/WASM isolation)
- [ ] Web dashboard with tool marketplace
- [ ] Tool versioning + rollback
- [ ] Global “Dynamo Hub” for shared tools
---
## 🤝 Contributing
We welcome your PRs! 🚀 Build new core features, improve security, or design amazing logos/UX.
1. Fork it
2. Create feature branch (`git checkout -b feature/awesome`)
3. Commit (`git commit -m "Add awesome feature"`)
4. Push & PR
---
## 📜 License
MIT License © 2025 Talha Yousaf
---
## 💡 Tagline
> **Dynamo MCP — The last MCP server you’ll ever need.**
MCP Store
