youtube-video-research

Extracts quotes and data from YouTube videos to build citation-rich research reports.

Updated Aug 28, 2026
One-click install
npx skills add https://github.com/AnderHonorato/Mem-rias-IA---Infinity --skill youtube-video-research-anderhonorato
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: youtube-video-research
Source: https://github.com/AnderHonorato/Mem-rias-IA---Infinity/tree/main/Manus/Skills/youtube-video-research
Command: npx skills add https://github.com/AnderHonorato/Mem-rias-IA---Infinity --skill youtube-video-research-anderhonorato

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Deep research based only on text articles misses first-hand evidence such as CEO interviews, earnings calls, expert panels, and conference talks. This Skill turns YouTube into a primary research source by orchestrating video discovery, structured video analysis, and synthesis into quote-rich, citation-backed reports. ## Core Features & Use Cases - Theme-Based Video Discovery: Decomposes a research goal into 4-6 themes and selects 2-3 videos per theme, targeting 12-20 videos for deep research with balanced bullish and critical perspectives. - High-Density Video Analysis: Uses manus-analyze-video with mandatory structured prompts that demand direct quotes, exhaustive data points, and sentiment ratings instead of thin summaries. - Quote-Rich Report Synthesis: Produces reports with mandatory blockquote citations (at least one per 300 words), expanded findings, multi-source triangulation, and a Source Videos table. - Use Case: For an investment analysis of a company, the Skill finds CEO interviews, earnings calls, and bearish analyst videos, extracts verbatim quotes and financial metrics from each, cross-validates claims with web search, and delivers a bull-versus-bear report with full source attribution. ## Quick Start Research the competitive landscape of the AI chip market using YouTube expert panels and CEO interviews, and produce a quote-rich report with cited video sources.

Frequently Asked Questions about youtube-video-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use YouTube videos as sources for deep research?

Decompose your research goal into 4-6 themes, search YouTube for 2-3 videos per theme, then analyze each video with manus-analyze-video using a structured prompt. Cross-validate claims with web search and synthesize findings into a report with blockquote citations.

What prompt should I use with manus-analyze-video for best results?

Use the high-density template that demands at least 8 direct quotes, all numbers and dates, and structured sections A through F including speaker profile, data points, arguments, risks, and sentiment. Generic summarize prompts produce thin, quote-free output.

How many YouTube videos should I analyze for a research report?

Target 12-20 videos for deep research and 8-12 for standard research, selecting 2-3 videos per theme. For investment or company research, ensure at least 30 percent of videos present critical or bearish perspectives to avoid confirmation bias.

Can I trust claims made by speakers in YouTube videos?

No, video content should not be treated as ground truth since speakers may exaggerate or present biased views. Cross-validate factual claims with web search and label opinions versus facts in the final report.

What happens if video analysis fails or a video is unavailable?

If a video is deleted, lacks a transcript, or is too long, use web search to find text summaries, articles, or alternative videos covering the same topic. The research process continues rather than halting on a single failure.