# Developer Zone

The **Developer Zone** is designed for builders who want to extend or integrate **myNeutron** into their own workflows, AI tools, or IDEs.

You can connect directly using the **Model Context Protocol (MCP)** today, or prepare for the upcoming **myNeutron API** for broader integration.

### **MCP Technical Overview**

**Model Context Protocol (MCP)** is an open, secure framework that lets AI tools retrieve structured data directly from external sources. With MCP, your IDEs and AI assistants can access your personal Seeds and Combined Contexts in real time.

#### Key Concepts

| Element           | Description                                                                       |
| ----------------- | --------------------------------------------------------------------------------- |
| **MCP Server**    | myNeutron’s backend endpoint that manages authentication and serves context data. |
| **MCP Client**    | The tool or AI (e.g., Claude, Cursor, VS Code) that requests your data.           |
| **Access Token**  | A personal key authorizing a specific client to access your Seeds.                |
| **Session Scope** | Determines which datasets (Seeds / Contexts) are available for each tool.         |

#### How It Works

1. You obtain your **Server URL** and **Access Key** from your dashboard.
2. You register myNeutron as a provider in your MCP-compatible tool.
3. The tool fetches your Seeds or Combined Contexts via secure HTTPS calls.
4. Queries or code suggestions are enriched with your personal memory.

> **Privacy First:** Only the authorized tool can access your Seeds. Data is end-to-end encrypted and never shared beyond your permission scope.

***

### **JSON Config Templates**

Each MCP client (Claude, Cursor, VS Code, etc.) requires a configuration snippet.\
Here’s the general template:

```json
{
  "provider": "myNeutron",
  "server": "https://api.myneutron.ai/mcp",
  "token": "YOUR_ACCESS_KEY",
  "description": "Access to myNeutron Seeds and Contexts"
}
```

#### Example – Claude Code

```json
{
  "mcp": {
    "servers": {
      "myNeutron": {
        "url": "https://api.myneutron.ai/mcp",
        "authToken": "YOUR_ACCESS_KEY"
      }
    }
  }
}
```

#### Example – Cursor Editor

```json
{
  "providers": [
    {
      "name": "myNeutron",
      "server": "https://api.myneutron.ai/mcp",
      "token": "YOUR_ACCESS_KEY"
    }
  ]
}
```

Once added, restart your tool. You should see “myNeutron Connected” in your integrations list.

***

### **Connecting to IDEs & AI Tools**

myNeutron works with any MCP-enabled environment.\
Below are tested setups and common use cases.

#### **Claude / Claude Code**

* Add myNeutron via **Settings → Integrations → MCP Servers**
* Paste your Server URL and Access Key
* Restart Claude to enable personal context querying

**Use Case:** Draft or summarize directly in Claude using your private Seeds (e.g., “Summarize my investor updates.”)

***

#### **Cursor**

* Open **Settings → MCP Connections**
* Add a new provider with your URL + Token
* Save and restart

**Use Case:** Enrich Cursor’s coding suggestions with your project documentation from myNeutron.

***

#### **VS Code**

1. Open the **Command Palette (Ctrl/Cmd + Shift + P)**
2. Search “MCP Settings”
3. Click “Add Provider” and enter your details
4. Restart VS Code

**Use Case:**\
Ask your AI pair-programming assistant to reference your Seeds while generating or reviewing code.

***

#### **Other MCP Clients**

Any agent or IDE supporting the MCP standard (e.g., Windsurf, LiteLLM, custom AI bots) can connect.

**Verification Query:**

```
GET /contexts/list
```

Expected Response → Your available Combined Contexts.

> **Note:** myNeutron can support both public (read-only) and private (authenticated) scopes depending on your API key type.

***

### **API (Planned)**

The upcoming **myNeutron API** will allow direct programmatic access to the same memory layer that MCP exposes. Ideal for advanced builders, data scientists, or AI agent developers.

#### Planned Endpoints

| Method   | Endpoint            | Description                       |
| -------- | ------------------- | --------------------------------- |
| `POST`   | `/seeds/create`     | Upload new Seeds programmatically |
| `GET`    | `/seeds/{id}`       | Retrieve a single Seed            |
| `POST`   | `/contexts/combine` | Create a Combined Context         |
| `GET`    | `/contexts/list`    | List all Combined Contexts        |
| `POST`   | `/query`            | Query a context directly          |
| `DELETE` | `/seeds/{id}`       | Remove a Seed                     |

#### Features Under Development

* OAuth2 / API Key authentication
* Token scoped permissions
* Webhooks for auto-sync with external sources
* Rate-limit handling and event streaming

> Developers can already test integrations via MCP. The REST API will offer identical logic but in HTTPS format.

***

### **Custom Integration Examples**

These examples show how you can extend myNeutron into your own systems.

#### 1. **Slack Bot Memory**

* Create a Slack bot that saves messages into myNeutron Seeds via the API.
* Let users query Slack history through myNeutron Assistant.

**Benefit:** Convert team conversations into searchable, structured knowledge.

***

#### 2. **AI Agent Integration**

* Build a personal agent (e.g., with LangChain or ElizaOS).
* Use MCP to pull user-specific Combined Contexts from myNeutron.
* The agent can then reason over that memory to generate personalized outputs.

***

#### 3. **Knowledge-Linked Docs**

* Connect internal company wiki pages to myNeutron.
* Every update auto-creates a new Seed version.
* Developers can query internal docs directly from IDEs.

***

#### 4. **Custom Dashboards**

* Use the API to display metrics like “Total Seeds Created” or “Top Context Topics.”
* Combine with analytics tools for internal reporting.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://neutron-4.gitbook.io/neutron-docs/getting-started/developer-zone.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.