Social media
Create visual stories, short videos and imagery around an idea your audience can connect with.
Explore Social mediaSearch YouTube videos and channels, read transcripts and comments, and measure performance with live data your agent reads through SandBase.

Search videos, analyze channels, extract captions, read comments, and discover trends on YouTube through SandBase. Use when asked for YouTube research, video discovery, channel analysis, content strategy research, comment sentiment, or video transcript extraction.

Extract full transcripts and captions from YouTube videos for content analysis, summarization, note-taking, and repurposing. Supports multiple languages and auto-generated captions. Ideal for research, content creation, and accessibility workflows.
The connector-fit theme has four comments, not four independent threads. Supplied transcripts do not explain that step.
Fictional example data. These describe structure, not wording; the next title and script should be written from the creator's own tests. Template 1 is supported by three videos; template 2 by one, so it is weaker.
Fictional example data. Tags: 4; counted characters 87 of 500 (3 commas included; +2 for each tag with a space). - hangboard training: 20 - hang board: 12 - fingerboard climbing: 22 - finger strength for climbing: 30


Before you organize every note, pick one question you want to answer. Open a note that helps with that question. Add a sentence explaining why it matters, then link to the next useful note. This is a small workflow to try, not a promise that one system fits everyone.
Concept: one joint diagnostic demonstration, “Cable or connector?” PatchBench contributes its existing cable-continuity demonstration; SocketSketch is invited to contribute a connector-fit demonstration.
You can now check the cable. Next, check connector fit before replacing the socket. Open the connector-fit video for a closer look at loose and firm connections.
The YouTube total of 300 already contains membership receipts of 50, so the membership component is not added again.
Start with one question, then connect the notes that help you answer it.
Fictional example data. The goal (two long videos and two Shorts) needs A + B + E = 19.0 h against 10.8 plannable hours, 8.2 h short. This week can finish one long video and two Shorts. B moves to next week; C after that. Spare: 4.0 h (3.0 on Saturday, 1.0 on Sunday), of which 2.7 h is the buffer. A is published on Saturday, so its work needs 7.5 of the 11.5 hours from Monday to Saturday.
Fictional example data. Change for the next upload. Open with the problem in the viewer's words, as pb-08 does, and move the greeting, shop news and reminders to after the first step (or to the end). If there is a sponsor, place the read after the first concrete step.
Fictional example data. - Likely cuts (C01, C02, C04, C08): 8.1 s. Raw 108 s becomes about 99.9 s. - Check-first rows (C03, C05, C07, C10, C11, C12): 20.8 s. Cutting all of them as well would give about 79.1 s, but C07 and C10 are probably demonstrations. - Kept: C06 (deliberate repetition, 1.5 s). C09 is inside C08.
Fictional sample data. From 2:07.0 to 2:38.5, retention falls 14.2 points, from 64.0% to 49.8%. That is 8.7 times the typical rate of decline in the body of this video. Relative retention falls from 0.55 (above typical for videos of this length) to 0.35 over the same span. The sheet-building resumes at 2:50.0 ("Okay, back to the sheet"), 48 seconds after the story began. The title promises a budget sheet in 10 minutes; the backstory does not move the sheet forward. Not established: Why viewers left. The transcript shows what was said, not what was on screen (for example whether the sheet stayed visible during the story) or the audio.

Analyze influencer performance metrics across TikTok, Instagram, YouTube, Bilibili, and Kuaishou. Compare engagement rates, audience growth, content consistency, and cross-platform presence for data-driven partnership decisions.

Discover and evaluate KOLs (Key Opinion Leaders) and creators across TikTok, Instagram, YouTube, and Xiaohongshu through SandBase. Use when asked for influencer research, creator outreach lists, KOL evaluation, or partnership prospecting.

Track thought leaders and industry voices across LinkedIn, Twitter, and YouTube. Monitor publishing cadence, content themes, engagement patterns, and audience growth for competitive positioning and relationship-building intelligence.

Research any topic across the last 30 days using Reddit, Twitter, YouTube, Hacker News, web search, and news sources through SandBase. Use when asked for recent trends, launch reactions, competitor activity, person/company profiles, or current event analysis.

Research podcasts, episodes, and audio content by searching YouTube (where most podcasts crosspost), web sources, and transcript databases. Discover relevant shows, track guest appearances, and analyze topic coverage in the podcast ecosystem.

Identify emerging trends early by monitoring signals across Twitter trending, TikTok viral content, YouTube rising videos, Google News velocity, and Reddit community buzz. Combines platform-specific trend indicators into a unified trend confidence score.

Generate content ideas backed by real audience demand data from Reddit questions, YouTube search gaps, Twitter discussions, and web search trends. Identifies proven topics, underserved questions, and content formats that outperform in your niche.
Search YouTube videos and channels, read transcripts and comments, and measure performance with live data your agent reads through SandBase.
Install the Skill in your agent, connect SandBase and name the videos, channels or topics to look up on YouTube.
Start with a task, then explore the possibilities.
Create visual stories, short videos and imagery around an idea your audience can connect with.
Explore Social mediaRun deep web, news and academic searches and turn the results into sourced research.
Explore Search & researchAnalyse keywords, search results, backlinks and site health to plan content and growth.
Explore SEO & growthCompare the inputs and deliverables before you begin.
| What you have | What to look for | Next step |
|---|---|---|
| A clear brief | Install the Skill in your agent, connect SandBase and name the videos, channels or topics to look up on YouTube. | Browse this collection |
| A connected agent | Install the Skill with its NPX command and connect your agent to SandBase. API calls are billed to your SandBase organization. | Read the Skills guide |
| A different output | Create visual stories, short videos and imagery around an idea your audience can connect with. | Social media |
Copy the NPX command from the Skill page and run it in SandBase Agent, Codex, Claude or Cursor.
Connect your agent to SandBase so the Skill can call its data APIs.
Describe what you need, then check the sources in the output before you use it.
A few things to know before you start.
A set of instructions your agent follows to call SandBase data APIs, such as search, social or business data, and turn the results into a sourced answer.
You need a SandBase account connected to your agent. SandBase API calls are billed to your SandBase organization.
SandBase Agent, Codex, Claude and Cursor. Each Skill page lists its supported agents and the install command.