Skip to main content

Hoox MCP Server

Official Model Context Protocol (MCP) server for Hoox — the AI video generation platform. This server lets any MCP-compatible client (Claude Desktop, Cursor, Windsurf, etc.) generate videos, manage avatars and voices, and export MP4 files using natural language.

Quickstart with Claude Desktop

  1. Get your API key from app.hoox.video → Settings → API Keys. An Enterprise plan is required to access the API.

  2. Install uv (Python package manager):

    macOS / Linux

    brew install uv
    

    Windows (PowerShell)

    irm https://astral.sh/uv/install.ps1 | iex
    

    If you run into issues, see the uv installation documentation.

  3. Open your claude_desktop_config.json (Claude → Preferences → Developer → Edit Config)
    and add the "mcpServers" block alongside your existing settings:

Before

{
  "preferences": {
    "quickEntryShortcut": "off",
    "coworkScheduledTasksEnabled": false,
    "sidebarMode": "chat"
  }
}

After

{
  "preferences": {
    "quickEntryShortcut": "off",
    "coworkScheduledTasksEnabled": false,
    "sidebarMode": "chat"
  },
  "mcpServers": {
    "Hoox": {
      "command": "uvx",
      "args": ["hoox-mcp@latest"],
      "env": {
        "HOOX_API_KEY": "hx_live_your_key_here"
      }
    }
  }
}

Add the "mcpServers" block to your existing config file — keep your other settings intact.

Other MCP clients

For Cursor, OpenClaw, Claude Code, or any other compatible client:

pip install hoox-mcp

To get the latest tools later, upgrade with pip install -U hoox-mcp and restart your client.

Then add the configuration to your client's MCP configuration file:

{
  "mcpServers": {
    "hoox": {
      "command": "hoox-mcp",
      "env": {
        "HOOX_API_KEY": "hx_live_your_key_here"
      }
    }
  }
}

Your MCP client can now interact with Hoox through the tools listed below.

Example prompts

⚠️ Hoox credits are required to use these tools.

Try for example:

  • "Use the Hoox API to generate a 60-second vertical product demo video for our new DTC skincare serum."
  • "List the male avatars available in an office setting"
  • "Create an avatar from this description: man in a conference room in a suit"
  • "Duplicate the video vid_xxx by changing the voice to a male English voice"
  • "Generate a 45 second script about our new DTC skincare serum, then start the video generation and export the result as MP4"
  • "Generate 4 product images using Flux 2 Pro with a 16:9 ratio"
  • "What's the cost to generate a 6-second video with Kling v3 Pro?"

Available tools

Voices

Tool Description
list_voices List available voices with optional filters (language, gender, tags)
get_voice Get details for a voice by ID

Avatars

Tool Description
list_avatars List available avatar looks with rich filters (gender, age_range, ethnicity, hair_color, place, action, emotion, accessories, selfie)
get_avatar Get full details for an avatar and its looks
get_avatar_look Get details for a specific look
create_avatar Create a new avatar from a prompt or reference images
edit_avatar Edit an existing look to create a new variation
get_avatar_status Check the creation status of an avatar

Asset generation

Tool Description
list_asset_models List AI models for image/video generation (filter by type, provider, tag)
get_asset_model Get full details + input schema for a specific model
get_asset_pricing Calculate the credit cost before starting a generation
create_folder Create an asset or video folder to organize the library (type: assets/videos)
list_folders List existing folders (optionally filter by type) to reuse their IDs
start_asset Start an image or video generation job (accepts folder_id, parent_media_id)
get_asset_status Check the status of an asset generation job

Tip — stay organized. When you work on a project or generate a batch of related assets, create a dedicated folder with create_folder (type: "assets") and pass its id as folder_id on each start_asset call. The assistant is primed to suggest this automatically. Reuse an existing folder via list_folders rather than duplicating.

Tip — keep the history. When a generation is derived from an existing asset (edit, animation/image-to-video, upscale, restyle, variation), set start_asset's parent_media_id to the source asset's id — in addition to passing it as the model reference. This links the output to its source in the dashboard's history tree so the full edit chain stays visible (and the new asset inherits the source's folder).

Script

Tool Description
generate_script Generate a video narration script from a prompt

Video generation

Tool Description
start_generation Start an AI video generation job
get_generation_status Check the status and progress of a generation job

Export

Tool Description
start_export Start exporting a generated video to MP4
get_export_status Check the status of an export job

Video management

Tool Description
duplicate_video Duplicate a completed video, optionally changing the voice or avatar

Typical workflow

Video generation

1. list_voices / list_avatars       →  pick a voice & avatar
2. generate_script (optional)       →  get narration text
3. start_generation                 →  get job_id
4. get_generation_status (polling)  →  get video_id
5. start_export                     →  get export job_id
6. get_export_status (polling)      →  get MP4 download URL

Standalone asset generation

1. list_asset_models                →  pick a model
2. get_asset_pricing (optional)     →  confirm credit cost
3. start_asset                      →  get asset_ids
4. get_asset_status (polling)       →  get output URL

Project / batch (recommended for several related assets)

1. list_folders                     →  reuse an existing folder, or
2. create_folder (type="assets")    →  get folder_id for the project
3. start_asset (folder_id=...)      →  repeat per asset, all grouped
4. get_asset_status (polling)       →  get output URLs

Configuration

Environment variable Required Default Description
HOOX_API_KEY Yes — Your Hoox API key (hx_live_...)

Troubleshooting

"HOOX_API_KEY is required"

Make sure the HOOX_API_KEY environment variable is set in your MCP client configuration. The key must start with hx_live_ or hx_.

Error plan_required (403)

Your Hoox account must have an Enterprise plan to use the API. Upgrade at app.hoox.video.

Error insufficient_credits (402)

Your workspace has run out of credits. Top up from the Hoox dashboard under Settings → Billing.

Error rate_limit_exceeded (429)

The API allows 100 requests per minute on the Enterprise plan. Wait a bit before trying again.

Tool not showing up in Claude Desktop / Cursor

  1. Check that hoox-mcp is installed: pip install hoox-mcp or uvx hoox-mcp --help
  2. Restart your MCP client after updating the configuration
  3. Check your client's logs for connection errors

Claude Desktop logs are located at:

  • macOS: ~/Library/Logs/Claude/mcp-server-hoox.log
  • Windows: %APPDATA%\\Claude\\logs\\mcp-server-hoox.log

Generation stuck in "processing"

Video generation can take between 1 and 5 minutes depending on duration and options. Use get_generation_status to monitor progress. If the job is stuck for more than 10 minutes, it may have failed — check the error field in the status response.

License

This project is licensed under the MIT License. See LICENSE for details.

Metadata

Release files for hoox-mcp 0.1.7

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for hoox-mcp 0.1.7
File Size Uploaded Provenance
hoox_mcp-0.1.7.tar.gz 59.8 kB With provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for hoox-mcp 0.1.7
File Interpreter ABI Platform
hoox_mcp-0.1.7-py3-none-any.whl Python 3 none any Details

Total release size: 75.3 kB

Release files / hoox_mcp-0.1.7.tar.gz

Download URL hoox_mcp-0.1.7.tar.gz
Size 59.8 kB
Tags Source
SHA-256 checksum
How to use checksums
d51d32bbffc5b79403ab39d0d01bcba0355e68c54ca8350359fdfa335b5b863d
BLAKE2b-256 checksum
How to use checksums
eac197cfa4afc2126a78dc27f5b02e729e7013e781b44fbefa0bbf6b87e92482
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 6, 2026.

Transparency log

Release files / hoox_mcp-0.1.7-py3-none-any.whl

Download URL hoox_mcp-0.1.7-py3-none-any.whl
Size 15.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
81f10c21cf7dc9a5d5e6fa33c4164f11810bad4b520291a57eeef2a9ef0494e4
BLAKE2b-256 checksum
How to use checksums
3ad26f659582cfc6d862aca0a6cb4c40796ccd627d1044e3d08dfaecb2d415d4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 6, 2026.

Transparency log

Security for hoox-mcp 0.1.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for hoox_mcp-0.1.7.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 6, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page