Skip to main content
Glama

YouTube Transcript & Search MCP Server

by com.transcriptout

Get transcript

get_transcript
Read-only

Get the transcript (subtitles) of a YouTube video. Accepts a video id or any YouTube URL. Set video_metadata=true to get the title, channel and duration in the SAME call — do not call get_video_info first, that is a second billed call for data this one already returns. format=text returns plain readable text (default, cheapest to read); format=json returns timed segments with start/duration in seconds; srt/vtt return ready subtitle file bodies and srv3 the raw YouTube XML. Prefer text unless you need timestamps or a file. Costs 1 credit.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
kindNoTrack kind. Omit to prefer a manual track and fall back to auto
langNoLanguage code of the track, e.g. 'en', 'de'. Default 'en'.
videoYesYouTube video id or URL
formatNo'text' = plain text (default), 'json' = timed segments, 'srt'/'vtt' = subtitle file body, 'srv3' = raw YouTube XML (srv3 does not combine with segment)
segmentNoMax characters per segment. Raise it when chunking the transcript for embeddings or retrieval — 500-1500 gives chunks with enough context; lower it for subtitle-sized lines. Left out, an auto-generated track is cut into ~180-character segments and a manual one is returned exactly as its author broke it, so pass this whenever you need one size regardless of which track answers.
video_metadataNoInclude the video's title, channel, duration and views alongside the transcript. Replaces a separate get_video_info call — same one credit either way.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
kindNo
languageNo
metadataNo
video_idNo
transcriptNoformat=json: array of {text, start, duration} segments (seconds); text/srt/vtt/srv3: one string in that format
available_langsNo

TDQS

A4.9/5.0
Behavior5/5

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

Adds meaningful behavior beyond annotations: states the tool costs 1 credit, is a billed call, and describes track/segment defaults and format outputs. The readOnlyHint annotation is consistent with this being a retrieval operation, and the description enriches trust with cost and default behavior details.

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

Conciseness5/5

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

Every sentence is purposeful and front-loaded with the core purpose. It packs format semantics, cost, and sibling-tool routing into a compact, readable description without redundancy.

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?

Given the rich schema, output schema, and read-only annotations, the description covers all the decision-relevant context: when to use, which format to choose, cost implications, and how to avoid duplicate calls. The agent has enough to invoke correctly without further research.

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

Parameters4/5

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

The input schema already covers all parameters with 100% description coverage, so the baseline is strong. The description adds practical selection guidance for format (e.g., 'cheapest to read') and clarifies the billing benefit of video_metadata in one call, going slightly beyond the schema.

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?

States a specific verb and resource: get the transcript of a YouTube video, and clarifies accepted input forms (video id or URL). It distinguishes itself from sibling get_video_info by explicitly noting that metadata is included and a separate call is unnecessary.

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

Usage Guidelines5/5

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

Provides explicit usage routing: 'Prefer text unless you need timestamps or a file' and warns against calling get_video_info first when video_metadata=true. This gives the agent clear decision rules for format and metadata usage.

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

All YouTube Transcript & Search MCP Server toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.1/5.0
Disambiguation4/5

Most tools target a distinct resource and action: search, list, metadata, transcript, job lifecycle, and credits. The get_transcripts_job / get_transcripts_result / get_transcripts_results trio is easy to blur at first glance, but the descriptions clearly differentiate job status, one result, and paged results.

Naming Consistency4/5

The set mostly follows a predictable verb_noun pattern: get_, list_, search_, submit_, cancel_. The main deviation is latest_channel_videos, which uses an adjective instead of a verb, and the result/results pair is mildly confusing, but overall naming is coherent and easy to navigate.

Tool Count5/5

14 tools is well within the sweet spot for a server with this scope. Each tool serves a clear niche: video lookup, metadata, single transcripts, batch jobs, and search are all represented without unnecessary or redundant tools.

Completeness5/5

The server provides full coverage of the transcript and search workflow: search YouTube, list channel/playlist videos, fetch or batch-fetch transcripts, check job progress/credits, and read results. There are no obvious missing operations that would force an agent into a dead end.

Resources