summarize
Summarize or extract text/transcripts from URLs, podcasts, and local files.
Best use case
summarize is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Summarize or extract text/transcripts from URLs, podcasts, and local files.
Teams using summarize should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/summarize/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How summarize Compares
| Feature / Agent | summarize | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Summarize or extract text/transcripts from URLs, podcasts, and local files.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Summarize
Fast CLI to summarize URLs, local files, and YouTube links using the `summarize` CLI tool.
## When to use (trigger phrases)
Use this skill immediately when the user asks any of:
- "use summarize.sh"
- "what's this link/video about?"
- "summarize this URL/article"
- "transcribe this YouTube/video"
- "what does this article say?"
## Prerequisites
Install the summarize CLI:
```bash
brew install steipete/tap/summarize
```
## Quick start
```bash
# Summarize a URL
summarize "https://example.com" --model google/gemini-3-flash-preview
# Summarize a local file
summarize "/path/to/file.pdf" --model google/gemini-3-flash-preview
# Summarize a YouTube video
summarize "https://youtu.be/dQw4w9WgXcQ" --youtube auto
```
## YouTube: summary vs transcript
For best-effort transcript extraction (URLs only):
```bash
summarize "https://youtu.be/dQw4w9WgXcQ" --youtube auto --extract-only
```
If the user asked for a transcript but it's huge, return a tight summary first, then ask which section/time range to expand.
## Model + API keys
Set the API key for your chosen provider:
- OpenAI: `OPENAI_API_KEY`
- Anthropic: `ANTHROPIC_API_KEY`
- xAI: `XAI_API_KEY`
- Google: `GEMINI_API_KEY` (aliases: `GOOGLE_GENERATIVE_AI_API_KEY`, `GOOGLE_API_KEY`)
Default model is `google/gemini-3-flash-preview` if none is set.
## Useful flags
| Flag | Description |
|------|-------------|
| `--length short\|medium\|long\|xl\|xxl\|<chars>` | Summary length |
| `--max-output-tokens <count>` | Max output tokens |
| `--extract-only` | Extract text only (URLs) |
| `--json` | Machine readable output |
| `--firecrawl auto\|off\|always` | Fallback extraction |
| `--youtube auto` | Apify fallback for YouTube |
## Fallback without CLI
If the `summarize` CLI is not installed, you can fallback to using the web tool:
```bash
# Use fetch_url to get content, then have the LLM summarize
curl -s "https://example.com" | head -c 50000
```
Then ask the LLM to summarize the fetched content.
## Config
Optional config file: `~/.summarize/config.json`
```json
{ "model": "openai/gpt-4o" }
```
Optional services:
- `FIRECRAWL_API_KEY` for blocked sites
- `APIFY_API_TOKEN` for YouTube fallbackRelated Skills
youtube-summarizer
Extract transcripts from YouTube videos and generate comprehensive, detailed summaries using intelligent analysis frameworks
summarize_text_with_key_details
Summarizes articles or text, presenting key information in a concise format. Adapts to prioritize completeness over brevity when requested, ensuring no critical details are lost.
bgo
Automates the complete Blender build-go workflow, from building and packaging your extension/add-on to removing old versions, installing, enabling, and launching Blender for quick testing and iteration.
u09955-decision-journal-maintenance-for-accessibility-services
Operate the "Decision Journal Maintenance for accessibility services" capability in production for accessibility services workflows. Use when mission execution explicitly requires this capability and outcomes must be reproducible, policy-gated, and handoff-ready.
u0225-oversight-uncertainty-communicator
Operate the "Oversight Uncertainty Communicator" capability in production for Human Oversight and Operator UX workflows. Use when mission execution explicitly requires this capability and outcomes must be reproducible, policy-gated, and handoff-ready.
u01482-constraint-compilation-for-healthcare-operations
Operate the "Constraint Compilation for healthcare operations" capability in production for healthcare operations workflows. Use when mission execution explicitly requires this capability and outcomes must be reproducible, policy-gated, and handoff-ready.
tzurot-council-mcp
Best practices for using the Council MCP server in Tzurot v3 development - When to consult external AI, how to structure prompts, model selection, and multi-turn conversations. Use when planning major changes or needing a second opinion.
typespec-m365-copilot-typespec-create-agent
Generate a complete TypeSpec declarative agent with instructions, capabilities, and conversation starters for Microsoft 365 Copilot Use when: the task directly matches typespec create agent responsibilities within plugin typespec-m365-copilot. Do not use when: a more specific framework or task-focused skill is clearly a better match.
typespec-create-agent
Generate a complete TypeSpec declarative agent with instructions, capabilities, and conversation starters for Microsoft 365 Copilot
type-inference-validation
Static type inference and validation for navigation paths
twitter-intel
Real-time X/Twitter intelligence - analyze accounts, track topics, and monitor keywords using live data. Use when you need current social media insights, competitor monitoring, or audience research.
twelve-data-automation
Automate Twelve Data tasks via Rube MCP (Composio). Always search tools first for current schemas.