MCP Video Gen
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| RNNOISE_REF | No | Pinned upstream commit for RNNoise. | |
| COMFYUI_HOST | No | Hostname of the optional ComfyUI backend. Typical value: host.docker.internal | |
| COMFYUI_PORT | No | Port of the optional ComfyUI backend. Default: 8188 | |
| CF_ACCESS_AUD | No | Cloudflare Access application audience tag. | |
| MAX_UPLOAD_MB | No | Maximum upload size in megabytes enforced while streaming files. | |
| PIPER_ENABLED | No | Whether Piper TTS is enabled. Default: false | |
| COMFYUI_SCHEME | No | HTTP scheme for ComfyUI. Default: http | |
| BLENDER_ENABLED | No | Whether to enable the optional Blender bridge. Default: false | |
| PUBLIC_BASE_URL | No | Public base URL for authenticated HTTP(S) file streams, e.g. https://your-public-host.example. Enables get_cached_file_resource to return /files/{file_id} URLs. | |
| RNNOISE_ENABLED | No | Whether RNNoise is enabled. Default: true | |
| WHISPER_CPP_REF | No | Pinned upstream ref for whisper.cpp. Default: v1.8.6 | |
| CF_ACCESS_VERIFY | No | Whether to verify Cloudflare Access JWT at the origin. Default: false | |
| RNNOISE_MODEL_URL | No | Public URL for the RNNoise model. | |
| VIDEO_MCP_VERSION | No | Release version or commit SHA to run. Default: latest | |
| WHISPER_MODEL_URL | No | Public URL for the Whisper model. | |
| BLENDER_BRIDGE_URL | No | URL of the authenticated Blender bridge. Default: http://host.docker.internal:9876 | |
| PIPER_PACKAGE_SPEC | No | Piper package specification. Default: piper-tts | |
| RNNOISE_SOURCE_URL | No | Public source archive URL for RNNoise. | |
| SILERO_VAD_ENABLED | No | Whether Silero VAD is enabled. Default: true | |
| WHISPER_MODEL_NAME | No | Whisper model name. Default: tiny-q5_1 | |
| COMFYUI_MODELS_PATH | No | Host path to ComfyUI models directory for optional read-only filesystem discovery. | |
| VIDEO_MCP_DATA_ROOT | No | Application runtime data root. Default: /data | |
| WHISPER_CPP_ENABLED | No | Whether whisper.cpp is enabled. Default: true | |
| BLENDER_BRIDGE_TOKEN | No | Bearer token used by the MCP server to authenticate with the Blender bridge. | |
| HYPERFRAMES_NPM_SPEC | No | HyperFrames npm package spec. Default: hyperframes@0.7.111 | |
| RNNOISE_MODEL_SHA256 | No | Expected SHA-256 of the RNNoise model. | |
| SILERO_VAD_MODEL_URL | No | Public URL for the Silero VAD model. | |
| WHISPER_MODEL_SHA256 | No | Expected SHA-256 of the Whisper model. | |
| CF_ACCESS_TEAM_DOMAIN | No | Cloudflare Access team domain, e.g. https://your-team.cloudflareaccess.com | |
| RNNOISE_SOURCE_SHA256 | No | Expected SHA-256 of the RNNoise source archive. | |
| WHISPER_CPP_BUILD_JOBS | No | Number of build jobs for whisper.cpp. Default: 2 | |
| HYPERFRAMES_SKIP_SKILLS | No | Whether to skip HyperFrames skills. Default: 1 | |
| SILERO_VAD_MODEL_SHA256 | No | Expected SHA-256 of the Silero VAD model. | |
| VIDEO_MCP_FORCE_REFRESH | No | Whether to force refresh the release source. Default: false | |
| CHATGPT_FILE_TIMEOUT_SEC | No | Timeout in seconds for ChatGPT attachment downloads. Default: 300 | |
| CLOUDFLARED_TUNNEL_TOKEN | No | Token used by the optional Cloudflare Tunnel sidecar. | |
| COMFYUI_CUSTOM_NODES_PATH | No | Host path to ComfyUI custom_nodes directory for optional read-only filesystem discovery. | |
| REMOTE_IMPORT_TIMEOUT_SEC | No | Timeout in seconds for remote imports. Default: 120 | |
| BLENDER_BRIDGE_TIMEOUT_SEC | No | Timeout in seconds for Blender bridge operations. Default: 7200 | |
| CHATGPT_FILE_MAX_REDIRECTS | No | Maximum number of redirects for ChatGPT attachment downloads. Default: 5 | |
| REMOTE_IMPORT_ALLOWED_HOSTS | No | Comma-separated list of allowed hosts for remote HTTPS imports. | |
| REMOTE_IMPORT_MAX_REDIRECTS | No | Maximum number of redirects for remote imports. Default: 5 | |
| WHISPER_MODEL_AUTO_DOWNLOAD | No | Whether to auto-download the Whisper model. Default: true | |
| CHATGPT_FILE_MAX_BATCH_FILES | No | Maximum number of files in a ChatGPT batch upload. Default: 20 | |
| VIDEO_MCP_CHECK_UPDATES_ON_START | No | Whether to check for updates on start. Default: true |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
This server cannot be deployed
Maintenance
ActivitySlowing
ResponsivenessNo issues