Follow Builders, Not Influencers

An AI-powered digest that tracks top AI builders on X and YouTube podcasts, remixes their content into digestible summaries. Follow builders, not influencers.

GitHub: github.com/zarazhangrui/follow-builders Stars: 6.3k | Forks: 838

What it is

A skill you install in your AI agent (OpenClaw or Claude Code) that delivers a daily or weekly digest of what top AI builders are saying — researchers, founders, PMs, engineers actually building things — as opposed to influencers regurgitating information.

The digest pulls from three source categories:

• Podcasts (6): Latent Space, Training Data, No Priors, White House AI, Tough Things, Equity
• X/Twitter builders (26): Includes Andrej Karpathy, Dylan Patel, Bloomberg AI, Zvi Mowshowitz, Cathie Wood, and others
• Official blogs (2): Anthropic Engineering, Claude Blog

Available in English, Chinese, or bilingual. Delivered to Telegram, Discord, WhatsApp, email, or in-chat.

How it works

1. A central feed is updated daily with the latest content from all sources (blog articles via web scraping, YouTube transcripts via Supadata, X/Twitter via official API)
2. Your agent fetches the feed — one HTTP request, no API keys
3. Your agent remixes the raw content into a digestible summary using your preferences
4. The digest is delivered to your messaging app (or shown in-chat)

No API keys needed. All content is fetched centrally. If you use Telegram/email delivery, those keys are stored locally in ~/.follow-builders/.env. Your configuration, preferences, and reading history stay on your machine.

Installation

Claude Code:

git clone https://github.com/zarazhangrui/follow-builders.git ~/.claude/skills/follow-builders
cd ~/.claude/skills/follow-builders/scripts && npm install

Then tell your agent: "set up follow builders" or invoke /follow-builders.

OpenClaw:

git clone https://github.com/zarazhangrui/follow-builders.git ~/skills/follow-builders
cd ~/skills/follow-builders/scripts && npm install

Customization

The skill uses plain-English prompt files to control how content is summarized:

• summarize-podcast.md — how podcast episodes are summarized
• summarize-tweets.md — how X/Twitter posts are summarized
• summarize-blogs.md — how blog posts are summarized
• digest-intro.md — the overall digest format and tone
• translate.md — how English content is translated to Chinese

Changes via conversation ("make summaries more concise," "focus on actionable insights") or direct editing. Take effect on the next digest.

Why interesting

Fills a real gap: keeping up with AI builder content is hard because it's scattered across X, YouTube, and company blogs. The "follow builders, not influencers" framing is a useful heuristic — and having an agent do the curation and summarization means you get a focused digest rather than a firehose.

The central-feed architecture is also interesting: one HTTP request, no API keys for content, all the heavy lifting (scraping, transcript extraction, summarization) happens upstream. The skill is effectively a thin client.

License

MIT

See also

• ai-editing — Using AI to edit Markdown notes in Obsidian (Claude Code, Codex)
• coding-agents — broader category