Skip to main content
Glama

glim.sh

by sh.glim
OverviewDiscussionsAdmin

YouTube Transcript

glim_youtube_get
Read-onlyIdempotent

Fetch a YouTube video transcript from a video URL or 11-char id. The transcript is cleaned server-side: deduplicated, tags/HTML stripped, with coarse [m:ss] timestamps - roughly a tenth the size of the raw captions. Default format='text' returns it inline (when it fits ~40K chars / ~10K tokens) so a single call gives you the text directly; long-form videos fall back to a download_url note. Pass format='json' for the same transcript plus transcript metadata (video_id, canonical url, language, origin, size) and a presigned download_url - for batch/programmatic use. Default origin='uploader_provided' (human captions); falls back to 'auto_generated' automatically if missing (counts as 2 upstream calls). Cached 7 days server-side.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
refYesYouTube video URL or 11-char video id (e.g. https://youtu.be/dQw4w9WgXcQ, https://www.youtube.com/watch?v=dQw4w9WgXcQ, or dQw4w9WgXcQ)
formatNoOutput format. 'text' (default): the cleaned transcript inline as plain text (omitted with a download_url note when it exceeds the ~40K-char inline cap). 'json': the same cleaned transcript plus transcript metadata (video_id, canonical url, language, origin, size) and a presigned download_url - for batch/programmatic use. Both formats return the identical cleaned, deduplicated transcript.text
originNo'uploader_provided' for human captions (default), 'auto_generated' for YouTube auto-captions.uploader_provided
language_codeNoISO 639-1 language code (e.g. 'en', 'de', 'fr')en

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • changedInput schema / properties / format / description
      Previous value: -"Output format. 'text' (default): the cleaned transcript inline as plain text (omitted with a download_url note when it exceeds the ~40K-char inline cap). 'json': the same cleaned transcript plus structured metadata and a presigned download_url - for batch/programmatic use. Both formats return the identical cleaned, deduplicated transcript."New value: +"Output format. 'text' (default): the cleaned transcript inline as plain text (omitted with a download_url note when it exceeds the ~40K-char inline cap). 'json': the same cleaned transcript plus transcript metadata (video_id, canonical url, language, origin, size) and a presigned download_url - for batch/programmatic use. Both formats return the identical cleaned, deduplicated transcript."
    • changedInput schema / required
      Previous value: -[
-  "ref",
-  "format"
-]New value: +[
+  "ref"
+]
  2. Changed1 schema field changed
    • changedInput schema / properties / format / description
      Previous value: -"Output format. 'text' (default): the transcript inline as plain text (omitted with a download_url note when it exceeds the ~80K-char inline cap). 'json': structured metadata + a presigned download_url for the WebVTT file, with no inline transcript - for batch/programmatic use."New value: +"Output format. 'text' (default): the cleaned transcript inline as plain text (omitted with a download_url note when it exceeds the ~40K-char inline cap). 'json': the same cleaned transcript plus structured metadata and a presigned download_url - for batch/programmatic use. Both formats return the identical cleaned, deduplicated transcript."
  3. First observed

TDQS

A4.7/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds significant context beyond annotations, including server-side cleaning (deduplication, stripping tags, timestamps), size reduction, inline vs download_url fallback for long videos, and origin fallback with double-API-call note. No contradiction with annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single paragraph is front-loaded with purpose. Every sentence adds value, but could be more scannable with bullet points. Still efficient and well-structured.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 4-parameter tool with 100% schema coverage and no output schema, description covers return values (inline text or json with metadata and download_url), caching behavior, and fallback logic. Very complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, and description adds meaningful context for each parameter: ref (URL or ID), format (inline vs json with size limits), origin (fallback behavior), language_code (ISO code). Goes beyond schema descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states it fetches a YouTube video transcript from a URL or ID, using specific verbs and resource. It distinguishes from sibling tools which are for other platforms (Amazon, GitHub, etc.).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides clear context on when to use (fetching transcripts) and details on formats, origins, and caching. Does not explicitly mention when not to use or alternatives, but given siblings are for different platforms, it is adequate.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All glim.sh toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources