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 DidierRLopes/get-y2b-clips --skill extract-y2b-insightsgit clone --depth 1 https://github.com/DidierRLopes/get-y2b-clipsWrote 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/didierrlopes/get-y2b-clips/extract-y2b-insights)<a href="https://agentmods.dev/skills/didierrlopes/get-y2b-clips/extract-y2b-insights"><img src="https://agentmods.dev/badge/skills/didierrlopes/get-y2b-clips/extract-y2b-insights/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/didierrlopes/get-y2b-clips/extract-y2b-insights"><img src="https://agentmods.dev/badge/skills/didierrlopes/get-y2b-clips/extract-y2b-insights.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 | $0.00075 | $0.02156 |
| Opus 5 | $0.00037 | $0.01078 |
| Sonnet 5 | $0.00015 | $0.00431 |
| Haiku 4.5 | $0.00007 | $0.00216 |
Grade A, and why
extract-y2b-insights 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 11d 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 — 232 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Extract YouTube Insights
Pull the highest-signal ideas out of a YouTube video — specifically the most controversial takes and the ones that read as genuinely new (ideas you would NOT easily find already discussed across the web). Output them two ways:
- A clean
.txtreport saved to disk - The same content printed in the terminal
This is the text-only cousin of get-y2b-clips: it reuses that skill's transcript
download + VTT parsing, but produces no video clips and no subtitles — just
distilled insights.
When to Use This Skill
Activate when the user:
- Provides a YouTube URL and wants "insights", "key ideas", "hot takes", "takeaways"
- Wants the "most controversial" points, or "ideas that don't already exist" / "novel ideas"
- Wants a written summary of the thinking in a video, not clips of it
If the user wants video clips → use get-y2b-clips.
If the user wants subtitles burned into a video → use add-subtitles.
Dependencies Check
command -v yt-dlp || echo "MISSING: yt-dlp"
yt-dlp is the only hard dependency (used to fetch the transcript). Install:
# macOS
brew install yt-dlp
# Linux / universal
pip3 install yt-dlp
ffmpeg is NOT required for this skill (no media is produced).
Input Requirements
- Required: YouTube URL
- Optional:
- Number of insights (default: 5–8, scaled to video length)
- Focus: lean more "controversial" vs more "novel" (default: both)
- Whether to web-check novelty (default: yes, when WebSearch is available)
- Output path (default:
./insights/<date>_<slug>/insights.txt)
Workflow
Phase 1: Setup
VIDEO_URL="USER_PROVIDED_URL"
VIDEO_TITLE=$(yt-dlp --print "%(title)s" "$VIDEO_URL")
CHANNEL=$(yt-dlp --print "%(channel)s" "$VIDEO_URL")
DURATION=$(yt-dlp --print "%(duration)s" "$VIDEO_URL")
TIMESTAMP=$(date +"%Y-%m-%d_%H-%M-%S")
SLUG=$(echo "$VIDEO_TITLE" | tr '/:?*"<>|\\' '-' | tr '[:upper:]' '[:lower:]' | tr ' ' '-' | cut -c1-50)
OUT_DIR="./insights/${TIMESTAMP}_${SLUG}"
mkdir -p "$OUT_DIR"
echo "Video: $VIDEO_TITLE ($((DURATION / 60)) min)"
echo "Output: $OUT_DIR"
What ships with it
3 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.
- 11d ago First seen · 232 lines · 75 tokens per session scan A b7f3ee03ec8e
extract-y2b-insights is a skill published in the GitHub repository DidierRLopes/get-y2b-clips (6 stars, last pushed 2mo ago), licensed MIT. It adds 75 tokens to every session and 2,156 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
asset-spec
Generate per-asset visual specifications and AI generation prompts from GDDs, level docs, or character profiles. Produces structured spec files and updates the master asset manifest. Run after art bible and GDD/level design are approved, before production begins.
team-audio
Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.
uw-game-feel-integrator
Inject juice and game feel into gameplay code using the project's chosen middleware. Use when adding screen shake, tweens, particles, audio feedback, haptics, hitstop, or any form of "juice" to gameplay events. Triggers on requests like "add juice", "make this feel better", "add screen shake", "polish this feature"…
team-narrative
Orchestrate the narrative team: coordinates narrative-director, writer, world-builder, and level-designer to create cohesive story content, world lore, and narrative-driven level design.
fec-svg-animation
A workflow for creating or reviewing animated SVG graphics, which are scalable vector images such as icons, logos, illustrations, and diagrams. It covers CSS, SMIL, Framer Motion, GSAP, accessibility, and reduced-motion behavior.
recursive-decomposition
Handle tasks that exceed the context window by decomposing them: size and filter the input, chunk it, run recursive sub-agents on independent parts, verify on small windows, and synthesise programmatically, following the Recursive Language Models (RLM) research by Zhang, Kraska and Khattab (2025). Use when a task…