summarize

name	summarize
description	Transcribe and summarize videos from YouTube, local files, Google Drive, Dropbox, and social media platforms (TikTok, Instagram, Twitter/X, Reddit) using any OpenAI-compatible LLM. Use when asked to summarize, analyze, extract insights, fact-check, or create study materials from video content.
license	MIT
compatibility	Requires Python 3.7+, ffmpeg, and at least one OpenAI-compatible LLM API key in .env. Social media platforms (TikTok, Instagram, Twitter/X, Reddit, Facebook) are handled by yt-dlp by default; Cobalt is used as a fallback for unsupported URLs.
metadata	{"author":"summarize","version":"1.0"}
allowed-tools	Bash Read

Overview

Runs python -m summarizer to transcribe videos and summarize them using a configurable LLM. Supports 12 summary styles and multiple providers. For full documentation see martino.docs.buildwithfern.com.

Never fetch, scrape, or download video URLs directly (e.g. with WebFetch or curl). The CLI handles all downloading, transcription, and summarization internally.

Quick Start

Step 1 — Run the CLI:

python -m summarizer --source "VIDEO_URL"

The tool uses the default provider from summarizer.yaml. No extra flags needed for basic usage.

Step 2 — Read the output file:

The CLI prints the saved filename. Read it from summaries/:

[+] Saved watch_20260207_234533.md
# Read: summaries/watch_20260207_234533.md

Step 3 — Show the result to the user.

File Locations

All paths are relative to the project root (where setup.py is). The local config file is summarizer.yaml. Do NOT look in your home directory or the skill directory.

File	Location
Config	./summarizer.yaml
API keys	./.env
Prompt templates	./summarizer/prompts.json
Output summaries	./summaries/<filename>.md

CLI Reference

python -m summarizer [OPTIONS]

Required

Flag	Description
--source	One or more video URLs or file paths

Provider Options

Flag	Description	Default
--provider	Named provider from summarizer.yaml	default_provider from YAML
--base-url	API endpoint URL (overrides provider)	From YAML
--model	Model identifier (overrides provider)	From YAML
--api-key	API key (overrides .env auto-matching)	Auto from .env

Source Options

Flag	Description	Default
--type	YouTube Video, Video URL, Local File, Google Drive Video Link, Dropbox Video Link	YouTube Video
--force-download	Skip YouTube captions, download audio instead	False
--transcription	Cloud Whisper (Groq API) or Local Whisper	Cloud Whisper
--whisper-model	tiny, base, small, medium, large	tiny
--language	Language code or auto (uses first available caption track, lets Whisper detect language)	auto

Processing Options

Flag	Description	Default
--prompt-type	Summary style (see Styles below)	From YAML defaults
--chunk-size	Characters per text chunk	From YAML defaults
--parallel-calls	Concurrent API requests	30
--max-tokens	Max output tokens per chunk	4096

Output Options

Flag	Description	Default
--output-dir	Directory to save summaries	summaries
--output-format	markdown, json, or html	markdown
--no-save	Print to stdout only, don't save file	False
--visual	Send video directly to a video-capable model (skips transcription)	False
--verbose, -v	Detailed progress output	False

Config Options

Flag	Description
--config	Path to config file (default: auto-detect)
--no-config	Ignore config file, use CLI args only
--init-config	Generate example summarizer.yaml and exit

Summary Styles

Use with --prompt-type. Defined in summarizer/prompts.json.

Style	Purpose
Questions and answers	Q&A extraction from content
Summarization	Standard summary with title
Distill Wisdom	Ideas, quotes, and references extraction
DNA Extractor	Core truth distilled to 200 words max
Fact Checker	Claim verification with TRUE/FALSE/MISLEADING labels
Tutorial	Step-by-step instructions from content
Research	Deep analysis with broader context
Reflections	Philosophical extensions beyond what is said
Mermaid Diagram	Visual concept map in Mermaid.js syntax
Essay Writing in Paul Graham Style	250-word essay in Paul Graham's style
Only grammar correction with highlights	Grammar cleanup with bold highlights
Distill Wisdom with diagrams	Textual insights with optional Mermaid diagrams

Examples

# Basic YouTube summary
python -m summarizer --source "https://youtube.com/watch?v=VIDEO_ID"

# Specific provider
python -m summarizer --source "URL" --provider gemini

# Extract key insights
python -m summarizer --source "URL" --prompt-type "Distill Wisdom"

# Fact-check with Perplexity
python -m summarizer \
  --source "URL" \
  --base-url "https://api.perplexity.ai" \
  --model "sonar-pro" \
  --prompt-type "Fact Checker"

# Mermaid diagram from a lecture
python -m summarizer --source "URL" --provider gemini --prompt-type "Mermaid Diagram"

# Visual mode (sends video to a vision-capable model, skips transcription)
python -m summarizer --source "URL" --provider nvidia --visual

# Local file
python -m summarizer --type "Local File" --source "./lecture.mp4" --provider groq

# Batch process
python -m summarizer --source "URL1" "URL2" "URL3" --provider gemini

# No config file (all explicit)
python -m summarizer \
  --source "URL" \
  --base-url "https://openrouter.ai/api/v1" \
  --model "google/gemini-2.0-flash-exp:free" \
  --api-key "sk-or-v1-YOUR_KEY" \
  --prompt-type "Tutorial" \
  --no-config

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

Warnings

• Windows + --no-save: Do NOT use --no-save on Windows — Unicode output crashes stdout. Always let the tool save to file, then read it.
• Social media: TikTok, Instagram, Twitter/X, Reddit, and Facebook are handled by yt-dlp by default. Cobalt is only used as a fallback if yt-dlp fails or does not support the URL.
• Transcription fallback: If YouTube captions are unavailable, the tool automatically falls back to audio download + Whisper transcription.
• Cloud Whisper: Requires a Groq API key (free tier available).
• Use --verbose for detailed progress and debugging.