Python + AI Office Hours

This repo supports two parts of the weekly Python + AI Office Hours workflow using GitHub Copilot and agent skills: preparing the news roundup before office hours, then generating and publishing the recording-based write-up afterward. Agent-facing instructions are in AGENTS.md.

📺 See a video recording of this repo in action

User workflow

Before office hours

1. Add a folder under office-hours/ for the week's session, named with the date in YYYY_MM_DD format.
2. Run the /generate-weekly-news-roundup skill to generate the weekly news roundup webpage.
3. Review the roundup for accuracy, attribution, working links, and upcoming events before presenting it.

After office hours

1. Add raw.md to the week's folder with the YouTube recording URL and any supplemental resources, such as pasted Discord chat logs or weekly slide content.
2. Run the /generate-writeup skill, then review questions_answers.md for accuracy and formatting.
3. Run the /post-comments skill to publish each Q&A to the GitHub Discussion and create comments.md.
4. Run the /generate-youtube-description skill to create youtube_description.md with chapter timestamps.
5. Run the /generate-linkedin-post skill to create linkedin_post.md.

Each completed week should have a structure like this:

office-hours/
  YYYY_MM_DD/
    office-hours-news.html       # Pre-event browser roundup
    raw.md                       # Recording URL and supplemental resources
    transcript.md                # Timestamped recording transcript
    live_chat.md                 # YouTube live chat, when available
    questions_answers.md         # Q&A write-up with timestamps
    comments.md                  # GitHub Discussion comment links
    youtube_description.md       # YouTube description and chapters
    linkedin_post.md             # LinkedIn recap

Resources

• Weekly office hours recordings & Q&A
• Weekly live office hours