xSkill AI

You can connect Cursor to the xSkill AI MCP server to access powerful AI content generation features directly from your editor. This MCP server enables you to submit generation tasks, monitor progress, and manage credits through a simple HTTP API, so you can incorporate AI capabilities into your Cursor workflow with minimal setup.

How to use

Connect to the MCP server using a client that supports the MCP protocol over HTTP. You will authenticate with an API key and then submit tasks such as text-to-video, image generation, or TTS, and retrieve results. Use the provided endpoints to manage tasks, check balances, and obtain media URLs. The server supports both standard HTTP endpoints and quick start methods via a direct install flow.

How to install

Prerequisites you need before installation: a Cursor-enabled environment and access to MCP through HTTP. You will configure the MCP connection with the URL and authentication method shown in the setup steps.

Option A: HTTP URL with Authorization header

{
  "mcpServers": {
    "xskill-ai": {
      "type": "http",
      "url": "https://api.xskill.ai/api/v3/mcp-http",
      "headers": {
        "Authorization": "Bearer sk-your-api-key"
      },
      "args": []
    }
  }
}

Option B: HTTP URL with API key in query parameters

{
  "mcpServers": {
    "xskill-ai": {
      "type": "http",
      "url": "https://api.xskill.ai/api/v3/mcp-http?api_key=sk-your-api-key",
      "args": []
    }
  }
}

One-Click Install (Deeplink) you can use to initiate setup directly from Cursor. After installing, add your API key in Cursor Settings → Tools & MCP → xskill-ai.

Configuration and usage notes

• You can list available models, submit generation tasks, check balances, and retrieve results using the MCP endpoints. - The server supports a range of features including text-to-video, image-to-video, first/last frame control, omni-reference mode, and native audio sync. - Ensure your API key is kept secure and never committed to public sources.

Recommended starting point: configure the xskill-ai MCP server using the HTTP configuration option that best fits your workflow, then verify your connection by listing models and submitting a small test task.

Security and access

Use Bearer token authentication or pass the API key as a query parameter as shown in the installation options. Keep your API key secure and rotate it periodically. Do not expose the key in client-side code or public repositories.

Available tools

list_models

List available AI models by category to help you choose the right model for your task.

get_model_info

Retrieve model parameters and schema to understand inputs, outputs, and capabilities.

submit_task

Submit an asynchronous AI generation task and receive a task ID to poll for results.

get_task

Check the status of a previously submitted task and fetch results when ready.

list_tasks

List tasks with filtering options to monitor multiple generations.

get_balance

Check your remaining account credits and track usage.

daily_check_in

Perform daily check-in to earn bonus credits.

parse_video

Extract watermark-free video URLs from supported platforms.

upload_image

Upload an image to cloud storage for use in tasks.

list_voices

List available TTS voices for speech synthesis.

text_to_audio

Convert text to speech using selected voice models.

voice_design

Create or customize voices from natural language descriptions.

voice_clone

Clone a voice from an input audio sample.

upload_audio

Upload audio to be used for voice cloning or design.

sync_generate_image

Instant image generation without polling.

sync_generate_video

Instant video generation without polling.

list_sync_models

List models capable of synchronous generation.

list_points_packages

List available credit packages for purchasing credits.

create_payment_qrcode

Generate a QR code for credit payments.