An advanced utility for extracting transcripts and generating AI-driven structured summaries from Spotify, Apple Podcasts, and RSS feeds.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install podcast-summarizer
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install podcast-summarizer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Podcast Summarizer?
The Podcast Summarizer is a powerful integration designed to bridge the gap between spoken audio content and actionable textual insights. By leveraging modern speech-to-text models and large language models (LLMs), it automates the tedious task of listening to hours of audio, delivering high-quality transcripts and concise, structured summaries instead. As one of the core ecosystem extensions, this tool expands the capabilities of Openclaw Skills by adding native media intelligence directly to your workflow.
Whether you are managing massive content repositories, researching industry trends, or building a personal knowledge base, this skill handles the heavy lifting of audio downloading, platform scraping, decoding, transcription, and prompt-driven summarization. It is specifically built to seamlessly parse complex sources like Spotify pages, Apple Podcasts, and dedicated RSS feeds, saving you hundreds of hours of manual labor.
Podcast Summarizer Use Cases
- Content Curation & Repurposing: Instantly turn long-form podcast episodes into written newsletters, social media posts, or blog articles.
- Rapid Research & Learning: Skim through dense educational or technical podcasts by generating long, medium, or short summaries.
- Accessibility Compliance: Automatically generate accurate transcripts for audio-only content using localized Whisper transcription settings.
- Automated Media Archiving: Build a searchable database of your favorite shows by feeding structured summaries into your knowledge management systems using Openclaw Skills.
How Podcast Summarizer Works
- Platform Resolution & Audio Fetching: The skill identifies the input source. For Spotify or Apple Podcasts, it scrapes the episode metadata and performs an RSS lookup to locate the original media file. For direct RSS feeds or MP3 URLs, it fetches the file directly.
- Audio Downloading: The underlying script securely downloads the target podcast episode or segment in MP3 format.
- Local Speech-to-Text Transcription: Using the Whisper CLI, the downloaded audio is transcribed into text with automatic language detection or a user-defined language override (e.g., Traditional Chinese).
- LLM-Based Summarization: The generated transcript is passed to a powerful LLM (using either your OpenAI or Gemini keys) to extract key insights, structured themes, and core takeaways.
- Structured Output: The summary and transcript are optionally written to a specified output file, ready for consumption.
Podcast Summarizer Setup
Prerequisites
Make sure you have the following CLI utilities and environment variables configured before using this component of Openclaw Skills:
- Whisper CLI: Install OpenAI's Whisper on your system for local transcription.
- API Keys: Export your LLM provider keys to your environment:
export OPENAI_API_KEY="your-openai-api-key"
# OR
export GEMINI_API_KEY="your-gemini-api-key"
Usage Commands
Invoke the summarizer script using the following CLI arguments:
# Summarize a Spotify episode directly (resolves RSS behind the scenes)
python3 {baseDir}/scripts/summarize_podcast.py "https://open.spotify.com/episode/xxx"
# Summarize a specific episode from an RSS feed
python3 {baseDir}/scripts/summarize_podcast.py "https://example.com/feed.xml" --episode 1
# Summarize a direct audio URL
python3 {baseDir}/scripts/summarize_podcast.py "https://example.com/episode.mp3"
Parameter Options
Customize the extraction and output using the CLI flags:
--language [lang]: Specify target language (e.g.,zhfor Chinese, default: auto).--length [short|medium|long]: Adjust the detailed depth of the summary.--transcript-only: Skip the LLM step and output the raw text transcript.--output [FILE]: Direct the outputs into a specific text file.
Podcast Summarizer Data Schema & Taxonomy
This skill processes external feeds and raw audio files, generating structured outputs. It can handle custom podcast structures such as regional hosting systems (e.g., SoundOn Taiwan) by resolving feed redirects. Below is how data streams through the utility:
| Data Component | Format / Source | Description |
|---|---|---|
| Input URL | Spotify, Apple Podcasts, RSS, or Direct MP3 | Primary ingestion link parsed to resolve audio files |
| Audio Buffer | Temporary Local MP3 File | Downloaded audio content passed directly to local Whisper instance |
| Transcript | Plain Text (.txt / console) | Generated speech-to-text string with timestamps |
| AI Summary | Structured Markdown | Final output structured by selected length (short/medium/long) and themes |
Regional RSS Handling (SoundOn Example)
To analyze specific SoundOn hosting URLs, base64 decode the feed_url parameter from player.soundon.fm to resolve structural mappings:
- RSS XML Feed:
https://feeds.soundon.fm/podcasts/{podcast-id}/soundon.xml - Direct VIP MP3:
https://rss.soundon.fm/rssf/{podcast-id}/feedurl/{episode-id}/rssFileVip.mp3
Podcast Summarizer Advanced Features
- Smart Spotify-to-RSS Scraper: Bypasses Spotify's closed ecosystem restrictions by resolving episode pages back to open-standard RSS feeds.
- Multi-LLM Engine Compatibility: Flexibly switch between OpenAI GPT models and Google Gemini engines depending on your active API keys.
- Granular Summary Customization: Control the structural length and depth of summaries, making it adaptable for quick indexes or deep-dive study documents.
- Localized Transcription Pipelines: Fully optimized to execute localized Whisper commands (
--language zh --model small), which is perfect for processing regional podcasts (such as Taiwanese SoundOn feeds) efficiently. - Seamless Integration with Openclaw Skills: Works in unison with other developer plugins to store transcripts, automate summaries on a cron schedule, or index audio data.
SKILL.md
Loading
Related Openclaw Skills
iOS MobPush Integration Skill
An automated engineering skill designed to integrate the MobTech MobPush SDK into iOS projects with minimal friction and maximum compliance.
GitHub Issue Writer Skill
The GitHub Issue Writer Skill provides developers with structured templates, safety checks, and repeatable workflows to draft clear, copy-paste-safe GitHub issues.
Quire CLI Integration
An AI-powered skill that interfaces directly with the Quire project management tool via its CLI to read and modify tasks, projects, comments, and workflows.
DimOS Robotics Development Skill
A comprehensive development tool for building, simulating, and debugging agentic robotics applications inside the DimOS ecosystem.
SEO Optimizer
An automated SEO optimization and auditing tool for HTML/CSS websites that identifies critical search issues, generates XML sitemaps, and implements structured data.
wisediag-checkup
Analyze and interpret medical checkup reports from PDFs or local files with WiseDiag's advanced AI health analysis.
Featured*
