Skip to main content
Glama

analyze_youtube

Read-only

Analyze public YouTube videos with Gemini to generate summaries, notes, tutorials, quotes, fact-checkable claims, or time-range answers.

Instructions

請 Gemini 觀看一支公開的 YouTube 影片,依 mode 產生分析,或回答 question。

回答使用繁體中文,附 [mm:ss] 時間戳,並區分【畫面】與【語音】來源。
只支援公開影片(私人、不公開影片無法分析)。

使用建議:
- 使用者沒有提供轉錄稿時,呼叫前先問一次要不要附上(步驟:影片頁面「顯示完整資訊」→「顯示轉錄稿」→ 全選複製
  → 貼回對話),或直接分析;使用者選直接分析就不要再問。轉錄稿要照抄使用者貼的文字放進 transcript;
  分段分析時只放該段時間的部分。
- 先用 mode="summary" 掌握全貌,再針對重要段落用 start/end(可搭配 question 或其他 mode)深入。
- 教學、實作影片要學做法:mode="tutorial";畫面上有程式碼、指令或小字時加 detail="default"。
- 長影片(超過約 20 分鐘)或要 notes、tutorial 這類長輸出時,用 start/end 分段,每段 5–10 分鐘。
- 結果最後會註明實際使用的模型,以及它是這個使用入口的「主要模型」還是「備援模型」。
  chat 入口(Claude Desktop 一般對話、網頁版、手機)因為約 80 秒的等待限制,主要模型就是較快的 flash-lite,
  不是額度或錯誤造成的;需要更深入的分析,建議改在 Claude Code 使用(主要模型較強、時間較長)。
- 用了備援模型時內容較淺,最後一行會寫出原因(含 503 過載時自動重試 1 次的結果),重要細節可以稍後重試。
- 失敗訊息會區分:503 模型過載(約 1 分鐘後再試)、504 Google 端處理逾時(用 start/end 縮短範圍)、
  本機等待逾時(chat 入口可改在 Claude Code 分析)。
- 結果最後一行若寫「今日額度已用完」,代表該模型要等每日重置(台灣時間約 16:00)才會再使用。

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "url": "<url>"
    },
    "name": "analyze_youtube"
  }
}

Input Schema

NameRequiredDescriptionDefault
endNo只分析片段時的結束秒數,必須大於 start。
urlYes公開 YouTube 影片網址:youtube.com/watch?v=、youtu.be/、youtube.com/shorts/ 或 m.youtube.com/watch?v=
modeNo分析模式。summary(預設):結構化摘要;notes:依時間分段的詳細筆記;tutorial:可照做的教學步驟,畫面上的指令、程式碼、提示詞逐字抄寫;quotes:重要原話;claims:影片中的事實性主張與講者給的證據。summary
startNo只分析片段時的開始秒數(非負整數)。
detailNo畫面解析度。low(預設)省 token;需要看清楚畫面細節或小字時用 default。low
questionNo要問影片的問題或想聚焦的範圍(最多 2000 字元)。mode=summary 時只回答這個問題;其他 mode 時照該模式的格式並聚焦在這裡。
transcriptNo使用者自己從 YouTube「顯示轉錄稿」複製貼上的轉錄稿(含時間戳),最多 120,000 字元。原封不動照抄使用者貼的文字,不要自己編寫、摘要或翻譯;使用者沒提供就不要填。有 start/end 時只放那段時間(前後多留約 30 秒)的轉錄稿即可,不要每段都放整份;server 也會依時間戳自動裁切。用來校正專有名詞拼法、讓 quotes 逐字引用。

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.6.0

TDQS

A4.8/5.0
Behavior5/5

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

Annotations already declare readOnly/openWorld/non-idempotent, so the safety bar is low, yet the description still adds substantial behavior the agent cannot get elsewhere: which model is primary vs fallback per entry point and why, that a fallback result is shallower and names its reason (including automatic retry on 503), that failures are differentiated into 503/504/local-timeout with distinct remedies, and that a 'quota exhausted' line means waiting for the ~16:00 Taiwan reset. Nothing here contradicts the read-only, non-destructive 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?

Purpose, output format, and constraints are front-loaded in the first three lines, and the rest is organized into a bulleted usage block plus failure-mode bullets, so it is scannable rather than a wall of text. It is long, and the model/entry-point explanation is somewhat redundant with the failure-mode bullet, but nearly every line maps to a decision the calling agent must make.

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 7-parameter, multi-mode tool with no output schema, the description covers everything an agent needs: transcript handling, mode selection strategy, segmentation thresholds, resolution trade-offs, output language/format, model fallback consequences, and per-error-code recovery. No material gap remains for correct invocation.

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?

Schema coverage is 100% and the schema descriptions are already detailed, so the baseline would be 3; the description earns extra credit by adding cross-parameter behavior the schema does not state — that start/end can be combined with question or any mode for deeper dives, that transcript must be copied verbatim (never summarized or translated) and only the relevant 5–10 minute slice plus ~30s padding should be pasted, and that detail="default" is warranted when code, commands, or small on-screen text appear.

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 opening sentence names a specific verb and resource ('analyze a public YouTube video with Gemini, by mode, or answer a question') and immediately scopes it ('only public videos supported; private/unlisted cannot be analyzed'). It also declares the output contract (Traditional Chinese, [mm:ss] timestamps, 【畫面】vs【語音】), so an agent knows exactly what kind of operation this is before reading the schema.

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?

The '使用建議' block gives explicit sequencing rules: ask once about a transcript before calling (with the exact copy-paste steps) or analyze directly and do not re-ask; start with mode="summary" then drill into segments via start/end; use mode="tutorial" for how-to videos; segment videos over ~20 minutes into 5–10 minute chunks. It also states when to switch surfaces (Claude Code vs chat entry) and when the fallback model means results will be shallower.

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

Deploy Server

Other Tools