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 agentmods add skills/kirvo-reporting/video-to-notes/codex-skillnpx skills add KIRVO-REPORTING/video-to-notes --skill codex-skillgit clone --depth 1 https://github.com/KIRVO-REPORTING/video-to-notesWrote 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/kirvo-reporting/video-to-notes/codex-skill)<a href="https://agentmods.dev/skills/kirvo-reporting/video-to-notes/codex-skill"><img src="https://agentmods.dev/badge/skills/kirvo-reporting/video-to-notes/codex-skill.svg" alt="Measured on agentmods" 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 | $0.00101 | $0.01236 |
| Opus 5 | $0.00051 | $0.00618 |
| Sonnet 5 | $0.00020 | $0.00247 |
| Haiku 4.5 | $0.00010 | $0.00124 |
Grade A, and why
video-to-notes 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 4d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video to Notes
Turn a supported video URL into a transcript-backed report. Prefer downloadable captions; use the configured local Whisper backend only when captions are missing or unusable.
Non-negotiable behavior
- Treat a message containing only a supported video URL as a complete request. Start the full report workflow immediately; never reply with “What do you want me to do with this link?”
- Treat YouTube watch, Shorts, live, and
youtu.beURLs equally. Preserve query parameters such ast,list, or tracking parameters when passing the URL to the processor. - When the user states a narrower goal such as “only download,” “only translate,” or “only extract subtitles,” perform that goal instead of forcing the full report workflow.
- Do not ask for cookies, credentials, language, or output destination before the first safe attempt. Ask only after a concrete failure proves user input is required.
- Respect
<workspace>/config.jsonpreferences. If no configuration exists, use local output and mention configuration only after delivering the result. - Read
metadata.jsonand the completetranscript.txtbefore writing claims about the video.
Bare URL workflow
-
Run
video-to-notes doctorand use the workspace path it returns. Do not assume the current repository or shell directory is the video workspace. If the command is missing or doctor fails, follow references/setup.md. -
Process into local artifacts first, regardless of the configured publishing target:
video-to-notes process "VIDEO_URL" --environment localThis prevents an incomplete report from being published before
summary.mdandtags.jsonexist. -
Read the returned
metadata.jsonand the entiretranscript.txt. -
Follow references/reporting.md to write grounded
summary.mdandtags.json. -
Finalize only after both files are ready. Select exactly one route:
- Local:
video-to-notes finalize "VIDEO_FOLDER" --environment local - Notion with CLI credentials:
video-to-notes finalize "VIDEO_FOLDER" --environment notion - Notion connector without CLI credentials: finalize with
--environment local, then follow references/notion.md. - Obsidian: follow references/obsidian.md.
- Local:
-
Return the primary reader-facing artifact: the Notion row URL, Obsidian note path/URI, or local
report.html. Include the local report as the backing archive when publishing remotely.
What ships with it
8 files 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.
- 4d ago First seen · 99 lines · 101 tokens per session scan A 8836caab1c06
video-to-notes is a skill published in the GitHub repository KIRVO-REPORTING/video-to-notes (106 stars, last pushed 14d ago), licensed MIT. It adds 101 tokens to every session and 1,236 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 2026-08-30.
Other skills, from other repositories
video-to-knowledge
Use by2kb to turn a Bilibili/YouTube URL or a downloaded local audio/video attachment into a transcript, a short abstract, and long-form study notes.
default-video-digest
Turn a transcript into grounded, structured notes for deep study.
book-to-skill
Converts books and documents (PDF, EPUB, DOCX, HTML, Markdown, plain text, RTF, MOBI/AZW with Calibre) into structured agent skills, extracting frameworks, mental models, principles, techniques, and anti-patterns. Use when the user wants to study a document through GitHub Copilot CLI, Amp, Claude Code, or Hermes…
mindos
MindOS: local knowledge assistant & shared KB. Keeps decisions, notes, SOPs, debugging lessons, research findings, preferences across sessions/agents. Core: save notes, search KB, organize files, run workflows, review, append CSV, hand off context, distill lessons. NOT for app source or paths outside KB. Triggers…
mindos-max
MindOS: local knowledge assistant & global memory layer. Keeps decisions, notes, SOPs, debugging lessons, architecture choices, research findings, preferences, conversation summaries for all connected agents. PROACTIVE: (1) search MindOS first for past context, (2) offer to save after valuable work, (3) persist key…
learning-notes-automation
学习笔记自动化：视频/播客转录 → 知识点提取 → 闪卡生成 → 知识图谱更新。触发词：学习笔记、闪卡、Anki、知识提取、视频学习.