vdl-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| VDL_URL | No | Base URL of your vdl instance | http://localhost:8000 |
| VDL_TOKEN | Yes | vdl MCP token (from the MCP tab) | |
| VDL_MCP_TIMEOUT | No | default request timeout (seconds) | 30 |
| VDL_MCP_READ_ONLY | No | Kill switch: expose only read tools (no download/transcribe/delete/bulk) | false |
| VDL_MCP_ALLOW_BULK | No | bulk actions | false |
| VDL_MCP_ALLOW_READ | No | resolve/list/status/get_transcript | true |
| VDL_MCP_TOOL_<name> | No | Per-tool override, e.g. VDL_MCP_TOOL_download=off — beats the group flag | |
| VDL_MCP_ALLOW_DELETE | No | delete files | false |
| VDL_MCP_ALLOW_DOWNLOAD | No | download | true |
| VDL_MCP_ALLOW_SETTINGS | No | read settings | false |
| VDL_MCP_ALLOW_TRANSCRIBE | No | transcribe | true |
| VDL_MCP_DOMAIN_ALLOWLIST | No | comma-separated host suffixes the assistant may download from |
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
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| resolveA | [read] Analyze a link and list the available media items and qualities — WITHOUT downloading. Use this to preview what a URL contains before calling download. |
| list_filesA | [read] List downloaded files in vdl. Optional filters: query (name/label substring), service (e.g. youtube), label. Returns name, size, duration, labels, in_feed. |
| job_statusA | [read] Status of a download job by id, or the list of all current jobs if no id is given. |
| get_transcriptA | [read] Return the transcript text (.txt) of a downloaded file, if it has been transcribed. If none exists yet, tells you to run transcribe(name) first. |
| downloadA | [writes] Download a video or audio from a URL into vdl (best quality).
|
| transcribeA | [writes] Transcribe a downloaded file locally with Whisper and return the text. 'name' is the downloads-relative path from list_files/download (e.g. 'youtube/clip.mp4'). Requires a Whisper model set in vdl (Settings → Transcription). Can take a while. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Each tool has a clearly distinct purpose: preview (resolve), download (download), monitor jobs (job_status), list files (list_files), transcribe (transcribe), and fetch transcript (get_transcript). No functional overlap.
All names use snake_case and are readable, but the pattern varies: verb_noun (list_files, get_transcript), verb (resolve, download, transcribe), and noun_noun (job_status). Mostly consistent but not a single pattern.
Six tools cover the full preview-download-monitor-list-transcribe workflow without being excessive. Each tool serves a necessary function in the pipeline.
Core lifecycle is fully covered: preview, download, job monitoring, file listing, transcription, and transcript retrieval. Minor gaps like deleting files or canceling jobs are missing, but the essential workflow is complete.