Pull a full week of transcripts from your curated research channels in a single API call. Run them through an AI summarizer. Walk into writing day with the raw material already prepared — no hours of video-watching required.
If your newsletter covers a technical field, you likely have a list of 10–20 YouTube channels you check each week — conference talks, expert interviews, product walkthroughs, community discussions. Watching them all is impractical. Skimming is unreliable. You miss things. You write based on what you happened to catch, not what was actually most worth sharing.
The transcript is the text version of every video — machine-readable, skimmable, and summarizable. But pulling transcripts manually for dozens of videos every week is its own time sink.
VidProxy supports two delivery modes. The webhook is for real-time push — it fires the moment a video is detected. For newsletter research, you often want the opposite: pull everything on a schedule and process it in a batch.
The Pull API is built for this. GET /api/videos?tag=AI+Research&since=7d returns every transcript from channels tagged "AI Research" over the past 7 days. One call. All your research. Ready to process.
Set up once. Runs every week without manual intervention.
Add all your curated YouTube channels to VidProxy. Label each subscription with a topic tag — "AI Research", "Market Analysis", "Dev Tools". VidProxy monitors them continuously and stores every new transcript as it's published.
Set a cron job or n8n schedule to run every Sunday. Call GET /api/videos?tag=AI+Research&since=7d. The response contains all transcripts published in the past 7 days, sorted by publish date. On Pro, AI Enrichment summaries are already included.
Pass the week's transcripts (or their pre-generated summaries) to a GPT prompt. Ask it to produce a ranked list of the most interesting ideas with one-paragraph explanations. Or generate draft newsletter sections directly. Your research context is built for you.
A script that pulls the week's transcripts, summarizes them, and sends a research digest to your writing workflow.
# Pull everything from the past 7 days for a tag curl "https://vidproxy.pro/api/videos?tag=AI+Research&since=7d" \ -H "Authorization: Bearer tr_live_xxxxxxxxxxxx" # Response structure { "videos": [ { "channel_name": "Two Minute Papers", "video_title": "The New Reasoning Model That Changes Everything", "published_at": "2026-04-28T14:00:00Z", "transcript_text": "Welcome back, dear fellow scholars...", "summary": "Breakdown of new reasoning benchmark results...", "key_takeaways": ["3x improvement on math benchmarks", "..."], "topics": ["ai", "reasoning", "llm"] } ], "query": { "tag": "AI Research", "since": "7d", "count": 23 } }
// weekly-digest.js — run every Sunday via cron const VIDPROXY_KEY = process.env.VIDPROXY_API_KEY; const TAGS = ['AI Research', 'Dev Tools', 'Market Analysis']; async function buildWeeklyDigest() { // Pull all tags in parallel const results = await Promise.all( TAGS.map(tag => fetch(`https://vidproxy.pro/api/videos?tag=${encodeURIComponent(tag)}&since=7d`, { headers: { Authorization: `Bearer ${VIDPROXY_KEY}` } }).then(r => r.json()) ) ); // Flatten and sort by publish date const videos = results .flatMap(r => r.videos) .sort((a, b) => new Date(b.published_at) - new Date(a.published_at)); // Build context for GPT — use pre-generated summaries if available const context = videos.map(v => `**${v.video_title}** (${v.channel_name})\n${v.summary || v.transcript_text.slice(0, 800)}` ).join('\n\n---\n\n'); // Generate digest const digest = await openai.chat.completions.create({ model: 'gpt-4o', messages: [{ role: 'user', content: `You are a research assistant for a tech newsletter. Below are summaries of YouTube videos published this week. Identify the 5 most interesting ideas worth featuring. For each, write a 2-3 sentence explanation of why it matters to a technical audience.\n\n${context}` }] }); // Send to Notion writing doc await pushToNotionDoc(digest.choices[0].message.content, videos.length); } buildWeeklyDigest();
The Pull API adapts to however your research process works.
Subscribe to conference channels (NeurIPS, WWDC, KubeCon). After a conference week, pull everything with since=14d and generate a "what happened at the conference" summary in minutes.
Subscribe to the investor relations channels of companies you cover. Pull transcripts quarterly with since=90d and generate comparative analysis across companies automatically.
Monitor channels that publish long-form expert interviews. Use keyword alerts to filter for guests who mention your topic area. Build a curated archive of relevant expert perspectives.
Use different subscription tags per newsletter vertical. One VidProxy account, multiple team members, each pulling their own tag. Agency plan supports unlimited channels and webhooks.
Many podcasts publish on YouTube. Subscribe to the YouTube channel instead of dealing with RSS audio feeds. Get searchable, full-text transcripts of every episode automatically.
Subscribe to university lecture series, research lab channels, and academic conference recordings. Build a searchable transcript archive for literature review work.
since=7d window returns all matching videos in one response.
Free tier is free forever. Add channels, start collecting, pull when you need them.
No credit card · Free tier forever