watchlisted
Install: claude install-skill vidnavigator/skills
# Watch a video
Usage: `/watch <link> [question]`. Without a question, summarize the video.
Get what the video says and what it shows from VidNavigator, then answer from both. Nothing is downloaded and no local tool is needed: no yt-dlp, no ffmpeg, no speech-to-text or vision model.
## Calling VidNavigator
Each step names a VidNavigator tool and its arguments. Call it in whichever way this session offers:
- **MCP tools.** If tools such as `watch_video` are in your tool list (their names may carry a prefix, for example `mcp__vidnavigator__watch_video`), call them directly. A tool that starts a background task returns a `task_id` and a `task_type`: call `get_task_status` with both until `task_status` is `completed` or `failed`. That call waits for the task, so make it again whenever it still answers `processing`.
- **The bundled CLI.** Otherwise run the script that sits beside this file, where `SKILL_DIR` is the directory containing this SKILL.md:
```bash
node "$SKILL_DIR/scripts/vn.mjs" watch_video video_url="https://www.youtube.com/watch?v=..." --transcript video.txt
```
Same tools, same arguments, written `key=value`. Quote a value that holds spaces, `&` or `?`. The CLI waits for background tasks itself. `--transcript FILE` writes what is said and what is shown as text, in the order of the video: a `[m:ss]` timestamp at each block, and `(picture)` on the lines that describe a scene. It also prints the video's details. Use it for every video, then read the file