# DeepSFetch MCP server

> Lets an AI client scrape web content, extract structured HTML data, and perform automated web searches using a hosted collection engine.

- Repository: https://github.com/ermultimidia/deepsfetch-developer-kit
- Page: https://hostedmcp.org/servers/ermultimidia/deepsfetch-developer-kit
- Transport: remote
- Stars: 4
- Latest version: 2.0.1

## Install

### Claude Code

```sh
claude mcp add --transport http deepsfetch-developer-kit https://mcp.deepsfetch.com/mcp
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "deepsfetch-developer-kit": {
      "url": "https://mcp.deepsfetch.com/mcp"
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "deepsfetch-developer-kit": {
      "type": "http",
      "url": "https://mcp.deepsfetch.com/mcp"
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"deepsfetch-developer-kit","type":"http","url":"https://mcp.deepsfetch.com/mcp"}'
```

### Codex

```sh
codex mcp add deepsfetch-developer-kit --url https://mcp.deepsfetch.com/mcp
```

### Codex config.toml (`~/.codex/config.toml`)

```toml
[mcp_servers.deepsfetch-developer-kit]
url = "https://mcp.deepsfetch.com/mcp"
```

### Gemini CLI

```sh
gemini mcp add -s user --transport http deepsfetch-developer-kit https://mcp.deepsfetch.com/mcp
```

### Gemini settings.json (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "deepsfetch-developer-kit": {
      "url": "https://mcp.deepsfetch.com/mcp",
      "type": "http"
    }
  }
}
```

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `DEEPSFETCH_API_KEY` | no | no |  |

## FAQ

### Does this repository include the collection engine?

No, the repository contains client-side tools like SDKs and a CLI, but the collection engine and hosted service are managed by DeepSFetch.

### What is required to run the provided examples?

Running the examples requires a DeepSFetch account and an API key, and collection usage consumes workspace credits.
