Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add andregusman-raiz/a-gusman-claude --skill youtube-transcriptgit clone --depth 1 https://github.com/andregusman-raiz/a-gusman-claudeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/andregusman-raiz/a-gusman-claude/youtube-transcript)<a href="https://agentmods.dev/skills/andregusman-raiz/a-gusman-claude/youtube-transcript"><img src="https://agentmods.dev/badge/skills/andregusman-raiz/a-gusman-claude/youtube-transcript/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/andregusman-raiz/a-gusman-claude/youtube-transcript"><img src="https://agentmods.dev/badge/skills/andregusman-raiz/a-gusman-claude/youtube-transcript.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00090 | $0.00651 |
| Opus 5 | $0.00045 | $0.00326 |
| Sonnet 5 | $0.00018 | $0.00130 |
| Haiku 4.5 | $0.00009 | $0.00065 |
Grade A, and why
youtube-transcript scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
YouTube Transcript — yt-dlp local
Converte video YouTube em transcript .md denso ANTES de ler — nunca WebFetch/browser na pagina do YouTube para extrair conteudo falado (nao funciona e desperdica tokens).
Origem: absorcao minima do gap identificado na avaliacao do Agent-Reach (2026-08-06, veredito P2/ignorar). Sem skill de terceiro no harness; so yt-dlp (ja instalado via brew).
Uso
bash ~/Claude/.claude/scripts/yt-transcript.sh <url-ou-video-id> [sub-langs]
# sub-langs default: "pt.*,pt,en.*,en"
- Output: caminho do
.mdem~/Claude/.cache/yt-transcripts/<video-id>.md(cache — 2a chamada e instantanea). - O
.mdtraz titulo, canal, data, duracao, idioma da legenda e o transcript limpo (sem timestamps, dedupe do rolling das auto-captions).
Fluxo canonico
- Rodar o script; capturar o path impresso.
- Grep/Read parcial no
.md— video longo (>30min) pode passar de 500 linhas; aplicar economia de tokens padrao (Grep antes de Read,offset/limit). - Sintetizar a resposta; citar o path do transcript.
Erros
| Exit | Significado | Acao |
|---|---|---|
| 2 | yt-dlp ausente | brew install yt-dlp |
| 3 | nao extraiu video id | conferir URL |
| 4 | video sem legenda (manual ou auto) | reportar ao usuario; fallback = baixar audio e transcrever local (sistema escuta / whisper Metal — ver memory gotcha_transcricao_arquivo_longo_ram_safe) |
| 1 + "HTTP Error 429" | throttle YouTube no endpoint de legendas (per-IP, agressivo) | aguardar 60-90s e re-tentar UMA vez; nao martelar. O script ja mitiga (player_client=android, sleep-subtitles 2, lista curta de langs — NUNCA wildcard pt.*, que casa tracks auto-traduzidos e multiplica requests) |
Limites
- So legendas ja existentes no YouTube (manuais ou auto-geradas). Video sem legenda => fallback whisper local, nunca silencioso.
- Read-only; nao baixa video/audio.
- Playlist: rodar 1 chamada por video (o script aceita 1 id por vez).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 42 lines · 90 tokens per session scan A 3960d63e31e0
youtube-transcript is a skill published in the GitHub repository andregusman-raiz/a-gusman-claude (19 stars, last pushed 4d ago), licensed MIT. It adds 90 tokens to every session and 651 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
docs-sync-internal
Use when code changes on the current branch need matching internal or developer documentation — "update our internal docs", "the architecture docs are stale after this change", "document what I just changed", "do the dev docs still match the code?" — or as a pre-push check that developer docs track the code. Narrower…
lab-report-writer
A writing workflow for producing structured, journal-style laboratory, research, engineering, or competition reports from experiment details and results.
community-profiler
A skill for analyzing technical community chat records from text, screenshots, JSON, or CSV. It creates member profiles, activity assessments, influence rankings, and community-health findings.
svg-flowchart
A tool for turning written steps into a consistent SVG flowchart. SVG is a scalable image format that can be embedded in web pages and documents.
frontend-dataviz
A skill for choosing and creating clear charts from user-provided data, following the Storytelling with Data approach. It matches chart types to tasks such as comparing categories, showing trends, or displaying distributions.
storybook-story-generation
Use after React components are built (Phase 4.5 of the Figma/Canva/screenshot pipeline) or whenever a project needs Storybook coverage — auto-generates .stories.tsx and .mdx docs from components via ts-morph AST parsing, with prop controls, variant stories, action args, and default args. Keywords: Storybook, generate…