Connect Your AI Clients to WolfHub

Seamlessly generate images, synthesize expressive voices, transcribe audio recordings, animate videos, and upload binary media from Cursor, Claude Desktop, ChatGPT, N8N, or custom scripts.

⚡ Client Quick Connection Setup

Need an API key? Get one on WolfHub.io →

Keys are never saved on the server. They are sent securely as Bearer tokens or query parameters to authenticate your client sessions directly with WolfHub.

Your Client SSE Endpoint URL

https://mcp.wolfhub.io/sse

📖 AI Client Connection Guides

1. Open Cursor Settings

Navigate to Cursor Settings (or press Ctrl+, / Cmd+,), then select Features → MCP Servers.

2. Add New MCP Server

Click + Add New MCP Server and fill in the details:

Name: wolfhub
Type: sse
URL:  https://mcp.wolfhub.io/sse

3. Verification

Click Add. A green status light will confirm your connection, and all 8 tools will immediately appear in your Composer and Chat windows.

Configure claude_desktop_config.json

Claude Desktop uses STDIO by default. To connect to this remote SSE server, use the lightweight mcp-remote bridge:

• Windows: %APPDATA%\Claude\claude_desktop_config.json
• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "wolfhub": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.wolfhub.io/sse"
      ]
    }
  }
}

Connecting with ChatGPT Actions or LibreChat

For platforms supporting Remote MCP over SSE:

Transport: SSE (Server-Sent Events)
Endpoint:  https://mcp.wolfhub.io/sse
Headers:   Authorization: Bearer YOUR_WOLFHUB_API_KEY

Connecting to N8N AI Workflows

In N8N, connect your AI Agent node to the MCP Tool node:

Select Server-Sent Events (SSE) transport mode.
Set Server URL to:

https://mcp.wolfhub.io/sse

If your N8N version uses HTTP headers for authentication, set Header Authorization to Bearer <YOUR_API_KEY>.

Test SSE Connection via cURL

You can verify live streaming from Linux, macOS, or Windows Terminal:

curl -N -H "Authorization: Bearer YOUR_WOLFHUB_API_KEY" "https://mcp.wolfhub.io/sse"

🛠️ Available MCP Tools

wolfhub_generate_image Images

Generates or edits images from prompts or reference images. Automatically polls and delivers CDN image URLs.

Default: z-image-turbo $0.00468 / img

wolfhub_generate_speech TTS

Converts text into realistic multilingual speech. Delivers direct high-quality WAV/MP3 audio URLs.

Default: eleven-multilingual-v2 ~$0.000048 / req

wolfhub_transcribe_audio STT

Transcribes speech to text with word-level timestamps and speaker diarization. Supports URLs or local files.

Default: grok-stt $0.0018 / min

wolfhub_generate_video Videos

Generates video scenes from text descriptions or animates reference images into video clips.

Default: seedance-2.0-pro (480p) ~$0.0106 / vid

wolfhub_upload_media Storage

Uploads local binary files, Base64 payloads, or remote URLs to WolfHub Share (up to 3GB, 7-day retention).

Endpoint: upload.wolfhub.io Free / Included

wolfhub_list_models Discovery

Discovers active models, unit types, providers, and prices across all 4 services without token bloat.

Services: images, videos, tts, stt Free

wolfhub_get_model_details Schema

Fetches live JSON Schema, constraints, allowed enums, and voice lists for any single model.

Returns: Live JSON Schema Free

wolfhub_check_task_status Polling

Manually checks task progress, cost, balance, and final media output for asynchronous jobs.

Status: processing/completed Free