Room Reader: an offline, open-weight read of the YouTube comments you never answered

Description and use

Room Reader is a single‑file Python tool that, without APIs or cloud services, identifies unanswered questions in a YouTube channel’s comments and outputs a ranked queue, enabling small‑channel owners to respond quickly to buyer inquiries.

Problem

Small YouTube channel owners need a quick way to identify unanswered questions in their comment sections without incurring API costs or exposing private data.

Outcome

In a test on four small wax‑melt channels, Room Reader scanned 90 comments, identified 41 for analysis, and produced a single unanswered question, demonstrating effective prioritization of buyer inquiries.

Workflow

• 1. Run python3 roomreader.py @handle --videos 5 --comments 60 --check-replies --json out.json
• 2. Room Reader fetches recent videos via YouTube innertube endpoints
• 3. It pulls comments from a public Invidious instance
• 4. For each comment, it runs the DeBERTa‑v3 NLI model to compute entailment probability for hypothesis "The viewer is asking a question."
• 5. It filters out comments with low probability or surface‑level false positives (e.g., presence of "?" or interrogative words)
• 6. It checks reply threads for the channel owner’s replies using the authorIsChannelOwner flag
• 7. It outputs a JSON file with a ranked list of unanswered question comments

Evidence

Room Reader is a single-file Python tool for the person who runs a small YouTube channel as a side hustle.

Source

It reads the last few videos, pulls the comments, and hands you back a short queue of the comments that still deserve an answer: the real questions, from real viewers, that nobody ever replied to.

Source

The heart of it is one function: QUESTION_HYPOTHESIS = "The viewer is asking a question."

Source

The reply-thread check changed how I see my own data. ... most flagged questions had in fact been answered this way.

Source