Skip to main content
Glama
trextrinorex

HyperFrames MCP Server

by trextrinorex

HyperFrames MCP Server

Remote MCP server for HyperFrames. It exposes HyperFrames project creation, file authoring, linting, validation, composition inspection, rendering, transcription, TTS, catalog access, documentation, and an advanced HyperFrames CLI tool over Streamable HTTP.

Architecture

Claude → Streamable HTTP MCP → this server → HyperFrames CLI → Chromium/FFmpeg → rendered video

This repository does not copy Nate Herk's student kit into your repository. It uses the public HyperFrames npm package directly. The student kit is MIT licensed, while its included AIS brand assets are not licensed for reuse; do not copy those assets into this server.

Local run

Requirements: Node.js 22+, FFmpeg, Chromium/Chrome.

npm install
npm start

The MCP endpoint is:

http://localhost:10000/mcp

Health check:

http://localhost:10000/health

Render deployment

This repo includes Dockerfile and render.yaml.

  1. In Render, create a Web Service from this GitHub repository.

  2. Select the Docker runtime (the Blueprint already specifies it).

  3. Deploy the main branch.

  4. After deployment, verify /health returns JSON with ok: true.

  5. Your MCP endpoint is:

https://YOUR-SERVICE.onrender.com/mcp

Render web services must listen on 0.0.0.0 and the PORT environment variable; this server does that by default. The included health check uses /health.

Claude MCP connector

Use the deployed HTTPS endpoint ending in /mcp as a remote Streamable HTTP MCP server.

Example:

https://YOUR-SERVICE.onrender.com/mcp

After connecting, Claude should discover tools such as:

  • hyperframes_init

  • project_write_file

  • project_read_file

  • project_list_files

  • hyperframes_lint

  • hyperframes_validate

  • hyperframes_compositions

  • hyperframes_render

  • hyperframes_transcribe

  • hyperframes_tts

  • hyperframes_catalog

  • hyperframes_add

  • hyperframes_docs

  • hyperframes_doctor

  • hyperframes_cli

Important storage limitation

The default Render filesystem is ephemeral. A project created during one running instance is not a permanent cloud drive and can disappear after instance replacement/redeploy. For serious production use, add persistent storage or object storage for project media and renders.

Security

The service is designed to be easy to connect first. If you expose it publicly, protect the MCP endpoint before using it with untrusted clients. Do not put private media, credentials, or secrets into project files.

HyperFrames

HyperFrames is an HTML-native video framework/CLI. The server uses the installed npm package and its CLI rather than attempting to reimplement the rendering engine.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/trextrinorex/hyperframes-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server