vidskiller

built a tool that watches youtube videos (not just the transcript) and extracts different types of structured output depending on what you need. 16 templates: code extraction, bias detection, playbook building, agent skill specs. free to try

@0xZane_ would love to see what you’re doing with it! thanks for connecting

@vidskiller Useful for content analysis

built a tool that watches youtube videos (not just the transcript) and extracts different types of structured output depending on what you need. 16 templates: code extraction, bias detection, playbook building, agent skill specs. free to try

@venturetwins lol incredible things happening

Real estate agents have discovered AI video and it’s glorious (from dawn.forkenbrock on IG)

@Codie_Sanchez Excellent point! Scary how good voice agents are getting

One of the most overlooked AI opportunities in the next 24 months is voice agents for boring businesses. Every HVAC, plumbing, and pest control company in America is sending calls to voicemail after 5pm. One AI voice agent fixes it overnight. Most owners have never even heard of this technology, yet alone know how to implement it. The person who packages this up and sells it to 500 of them is going to be very rich...very quietly.

@yashhq_22 very well put! a lot happening so quickly

AI agents are getting good enough to run your support, outreach, and onboarding. the scary part? most founders still aren’t using them.

@gregisenberg Great work, would love to see more of what you’re cooking up

In case you missed it... This 58 min video is the clearest introduction to AI agents, agent skills, md files, building AI employees on the internet and it's 100% free

@leap_xyz ❤️❤️😭

Dad I wish you were here to see this

KEY FINDING:
 Chamath admitted to a mistake in Nov 2021 of not de-risking his portfolio sufficiently, despite signals from "best capital allocators" like Musk and Bezos, because he confused "growing fame" with "growing skill," highlighting how even top investors succumb to ego.

ran a bias audit on a crypto video with 500K views findings... trust score came back 2.8/10 the information asymmetry in crypto YT is wild. built a tool to quantify it 😎

ran knowledge extraction on a trading video... the AI caught the speaker claiming their model 'smashed all benchmarks' while the on-screen data showed it lost in 5 out of 9 categories 🥴🤨🤔

the math on watching youtube tutorials average tutorial: 15 min pausing to copy commands: +10 min rewinding to catch config values: +8 min googling what you missed: +12 min total: 45 min for 15 min of content or: paste URL → 40 seconds → everything extracted, copy-paste ready

KEY FINDING: ◆ Emergence of "agentic payments" via platforms like Stripe, enabling AI agents to issue virtual cards for transactions (e.g., ordering lunch), signals direct financial integration of autonomous agents into real-world commerce, moving beyond mere info processing.

weird use case i didn't plan for but people keep doing: watching a strategy video where someone explains how they do something, then running Skill Seed on it it extracts the decision logic and YAML workflow specs, basically converts "how i do X" videos into agent skill cards

KEY FINDING: ◆ The most critical factor for YT success, often overlooked by creators, is selecting a topic with high "viral probability" (audience demand/curiosity triggers), as even perfect SEO, retention, and thumbnails are useless if nobody is searching for that topic.

most YT "AI summary" tools read the transcript the problem: in a 15 min tutorial, about 40% of the useful information is only shown on screen terminal commands, config files, folder structures, UI clicks, error messages none of that is in the transcript vidskiller solves it ❤️‍🔥

The Key Findings feature surfaces unexpected insights to the top of most analyses. Free to try at vidskiller now

◆ "Niche quilting" is highlighted as a highly effective new strategy where creators find viral video formats from other niches and adapt them to their own topic, as copying direct competitors is now a "path to nowhere."

KEY FINDING: ◆ YouTube's algorithm now prioritizes "satisfaction over retention," meaning it cares more about whether a viewer stays on the platform after your video, rather than just how long they watched your specific video. more YT insights below 📼❤️

Most YouTube creators are optimizing the wrong metrics in 2026. Watch time is dead. The algorithm now rewards Audience Retention Velocity in the first 15 seconds + comment velocity. I’ve seen channels 3x CTR by fixing just this. Who wants the framework? 👇