high-quality-blogs

High-Quality Technical Blogs & Sources

This is a curated registry of blogs and websites that produce textbook-quality technical content. When running the /research-textbook-chapter or /write-textbook-chapter workflows, the agent SHOULD search these sources for relevant material on the topic being researched.

How to Use This Registry

Attribution (MANDATORY)

These blogs contain original content by independent researchers and writers. When using their material in textbook chapters, you MUST:

1. Cite the author and post in the section's collapsible source header table
2. Attribute figures and diagrams in image captions: Source: [Author Name], "[Post Title]" ([year]). [URL]
3. Quote, don't paraphrase — when using a specific explanation or framing from a blog post, use a blockquote with attribution rather than rewording it without credit
4. Never present blog content as original — if an explanation or worked example is inspired by or adapted from a blog post, say so explicitly (e.g., "The following derivation is adapted from Gundersen's excellent treatment in [post title]")

During Research (MANDATORY)

When researching a topic, spend at least 2-3 searches specifically targeting these blogs:

site:lilianweng.github.io {TOPIC}
site:colah.github.io {TOPIC}
site:cameronrwolfe.substack.com {TOPIC}
site:magazine.sebastianraschka.com {TOPIC}
site:distill.pub {TOPIC}
site:gregorygundersen.com {TOPIC}

Pick the blogs most relevant to the topic's domain (see the category tags below).

During Source Downloading

Use the extraction method listed for each site. The authenticated_extract.py script handles most sites. Key notes:

• Substack sites auto-detect the article CSS selector
• GitHub Pages blogs usually work with default settings (no -s needed)
• Some sites need custom CSS selectors — see the "Extraction" field for each entry

Adding New Blogs to This Registry

When you discover a blog of exceptional quality during research, add it to this file following the template at the bottom. You MUST:

1. Read at least 3 posts from the blog to assess quality and depth
2. Test extraction with authenticated_extract.py to determine the best CSS selector
3. Categorize the blog using the depth tags defined below
4. List 2-3 notable posts with URLs
5. Add the entry in the correct category section