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 Ertinox7711/SGRR-AGI-V2 --skill instagram-reel-understandinggit clone --depth 1 https://github.com/Ertinox7711/SGRR-AGI-V2Wrote 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/ertinox7711/sgrr-agi-v2/instagram-reel-understanding)<a href="https://agentmods.dev/skills/ertinox7711/sgrr-agi-v2/instagram-reel-understanding"><img src="https://agentmods.dev/badge/skills/ertinox7711/sgrr-agi-v2/instagram-reel-understanding/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/ertinox7711/sgrr-agi-v2/instagram-reel-understanding"><img src="https://agentmods.dev/badge/skills/ertinox7711/sgrr-agi-v2/instagram-reel-understanding.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00191 | $0.01546 |
| Opus 5 | $0.00096 | $0.00773 |
| Sonnet 5 | $0.00038 | $0.00309 |
| Haiku 4.5 | $0.00019 | $0.00155 |
Grade A, and why
instagram-reel-understanding 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Instagram Reel Understanding
Goal: turn an Instagram URL into something you can reason about — full transcript (any language), key frames you can look at, and post metadata. After processing, hold all of it in context and wait for what the user wants to do with it (résumé, hooks, traduction, copy adaptation, fact-check, etc.).
When to run
Run as soon as you see an instagram.com/... URL in the user's message, OR when the user refers to an Instagram reel they just sent. If the user pastes several URLs, process them one after another (or in parallel if you have subagents and they're independent). Do not ask "voulez-vous que je le télécharge?" — that's the whole point of the skill, just do it.
Skip only if the user explicitly says "ne télécharge pas" / "don't fetch it".
How to run
Use the bundled script. It handles download, audio extraction, frame sampling, and Whisper transcription end-to-end. Use Python 3.11 — faster-whisper has prebuilt wheels there; Python 3.14 may fail to install.
& "C:\Users\YOU\AppData\Local\Programs\Python\Python311\python.exe" `
"C:\Users\YOU\.claude\skills\instagram-reel-understanding\scripts\process_reel.py" `
"<INSTAGRAM_URL>"
Defaults: model small (good multilingual quality, ~500MB one-time download), 8 key frames, output dir ./reel_output_<id> in the current working directory.
Useful overrides:
--model tiny→ fastest, lower quality. Use for very short reels or when the user wants speed.--model mediumor--model large-v3→ better quality, slower. Use when transcript fidelity really matters (legal, fact-check, careful translation).--frames 0(or--skip-frames) → audio-only, much faster. Use when the user only cares about what is said.--frames 16→ denser visual sampling for reels where the visual changes a lot.--out <dir>→ control where files land.
First run on a fresh machine will:
pip install faster-whisper(auto, ~200MB),- download the Whisper model weights (cached under
~/.cache/huggingface/). Both are one-time. Tell the user briefly so the wait isn't surprising.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago First seen · 85 lines · 191 tokens per session scan A fc6e0256af3e
instagram-reel-understanding is a skill published in the GitHub repository Ertinox7711/SGRR-AGI-V2 (1 stars, last pushed 4d ago), licensed MIT. It adds 191 tokens to every session and 1,546 once invoked, about $0.0010 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 2026-09-09.
Other skills, from other repositories
ag-testar-e2e
QA Playwright automatizado: navega como usuario, captura console/rede/UI quebrada e gera relatorio para fluxos criticos e regressao.
ag-capturar-tela
Captura tela e controla mouse/teclado em apps nativos macOS via router 3-tier (macos-use > macos-automator > computer-use). Para apps BROWSER, preferir Playwright MCP.
ag-testar-e2e-batch
Suite E2E completa com pre-flight de infra, batches de 10-20 testes, auto-fix de falhas (seletores/mocks/timing), retry ate 3 ciclos, progresso em /tmp/e2e-progress.md.
ag-referencia-playwright
Patterns canonicos Playwright 2026 — Chromium isolado, web-first assertions, locators semanticos, persistent context. Carregar antes de QAT/E2E/browser test.
ag-gerar-testes
Gera testes Playwright production-grade a partir de fluxos reais observados via MCP. Explora o app pelo browser, documenta cada passo, gera codigo TypeScript, valida executando. Use para transformar fluxos explorados em testes automatizados.
ag-testar-manual
Teste exploratorio com Playwright: navega como usuario, captura screenshots, console e a11y. Use antes de merge/deploy ou bugs visuais.