For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Get Started
  • Overview
    • Welcome
    • How It Works
  • Getting Started
    • Installation
    • Configuration
  • Usage
    • CLI Reference
    • Summary Styles
    • Batch Processing
    • Config Management
    • Retry Behavior
    • Errors and Troubleshooting
  • Features
    • Visual Mode
    • Transcription
    • Webapp
    • Caching
  • Integrations
    • Share a Summary
    • Cobalt
    • Proxy
    • Agent Skill
Get Started
On this page
  • Built-in Styles
  • Multi-Step Workflow
Usage

Summary Styles

Was this page helpful?
Edit this page
Previous

Batch Processing

Next
Built with

Use --prompt-type in the CLI or select from the dropdown in the web interface. Add custom styles by editing summarizer/prompts.json. Use {text} as the transcript placeholder.

Built-in Styles

StylePurpose
Questions and answersEssential questions that capture the main points, answered in detail
SummarizationStandard summary with a concise title
Distill WisdomKey insights, quotes, and references
DNA ExtractorCore truth distilled to 200 words max
Fact CheckerClaim verification with TRUE/FALSE/MISLEADING labels
TutorialStep-by-step instructions from the content
ResearchDeep analysis with broader context and connections
ReflectionsPhilosophical extensions beyond what is said
Mermaid DiagramVisual concept map in Mermaid.js syntax
Essay Writing in Paul Graham Style250-word essay in Paul Graham’s style
Only grammar correction with highlightsGrammar cleanup with bold highlights
Distill Wisdom with diagramsPrimarily textual insights with optional Mermaid diagrams

Multi-Step Workflow

For comprehensive analysis, chain multiple styles on the same video:

  1. Summarization — quick overview
  2. Distill Wisdom — extract key insights
  3. Fact Checker (with Perplexity) — verify claims
  4. Mermaid Diagram — visual reference

Transcripts are cached in memory, so re-summarizing the same source skips transcription entirely.