uoink-lite
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HTTPS_PROXY | No | Standard proxy environment variable. Set this in the server's config if you need a proxy to reach YouTube (for example when blocked or rate limited from cloud servers or VPNs). | |
| UOINK_LITE_HOME | No | Directory where fetched transcripts are saved as JSON files. Set this to store them somewhere other than the default location. | ~/.uoink-lite/transcripts/ |
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_transcriptA | Get a YouTube video's transcript, with [mm:ss] timestamps by default. Accepts any YouTube video URL (watch, youtu.be, shorts, embed) or an 11-character video id. lang is a caption language code such as "en" or "es". The transcript is saved locally so search_saved can find it later. |
| search_transcriptA | Find where a video talks about something. Returns matching transcript lines with timestamps and a deep link to each moment (https://youtu.be/ID?t=SECONDS). Matching is case-insensitive keyword matching, best matches first. |
| search_savedA | Search every transcript fetched before, across all saved videos. Returns the video, timestamp, matching line and a deep link for each hit, best matches first. Use this when the user refers to something they watched or asked about earlier. |
| list_savedB | List saved transcripts, newest first, with title, id and date. |
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 4 tools
The four tools have mostly distinct purposes: fetching a transcript, searching within a video transcript, searching across saved transcripts, and listing saved transcripts. The boundary between search_transcript and search_saved is somewhat subtle because both perform transcript searching, though the descriptions clarify one is per-video and the other is cross-video.
All names follow a consistent snake_case verb_object pattern: get_transcript, search_transcript, search_saved, list_saved. The convention is predictable and easy to scan.
Four tools is well-scoped for a lightweight YouTube transcript utility. Each tool covers a distinct part of the fetch-search-list workflow without unnecessary surface area.
The core lifecycle for fetching, searching, and listing saved transcripts is covered. Minor gaps exist around managing saved data, such as deleting or clearing saved transcripts, and possibly retrieving a full saved transcript without refetching.