two free sessions next week: LinkedIn content agent + why AI names your competitor

Upcoming free lessons

All free to attend. Register and show up.

Most LinkedIn content agents produce generic output because they don't know how you write. This session builds an agent in Claude that learns your tone, your topics, and your voice from a set of reference documents you create once. Give it an idea and it produces a LinkedIn draft that sounds like you wrote it, without post-editing to strip out the AI smell. You'll leave with the agent built and a voice document that makes every future draft progressively better.

→ Register now

You know AI is recommending your competitors in buyer queries. This session shows exactly why. Will breaks down what sources ChatGPT, Perplexity, Gemini, and Claude are actually using when they choose who to recommend in your category, and traces which specific pages are putting your competitor ahead of you right now. You'll leave with a clear picture of where the gap is and where to start closing it.

→ Register now

Featured articles from Catalyst

Three pieces from the blog worth reading this week.

The 4 Types of Pages That Drive AI Citations and How to Get Your Brand Featured on Them

82% of AI brand mentions come from pages you don't own, according to Geonimo's analysis of 50,000+ AI responses. This piece breaks down the 4 page types doing most of that work. Best-of lists account for 43.8% of ChatGPT citations (Ahrefs), followed by review platforms like G2, community threads (Perplexity pulls from Reddit about 47% of the time, per Profound's 30-million-citation study), and broad reference sources like Wikipedia. Brands with a Wikipedia entry get cited 4.1x more often. The fix starts with mapping which specific sources each model is using in your category right now.

→ Read the full article

How to Build a LinkedIn Lead Gen Agent That Qualifies ICP Prospects From Viral Posts

Pietro, founder of NotLinear Training and reaching about 500,000 people a month on LinkedIn and Substack, live-built this workflow at a Catalyst session. Paste any viral LinkedIn post in your niche and the agent scrapes the comment section using Apify, runs each profile against a plain-language ICP definition, and outputs a ranked list with a fit score and a personalized outreach message for each qualified lead. His demo extracted 82 commenters from a single post in under 30 minutes. Neither path (Relay App or Claude directly) requires connecting your personal LinkedIn account.

→ Read the full article

How to Build a Claude Workflow That Turns 1 Daily Idea Into 5 Ready-to-Post Drafts

Victoria Morales (Front Desk) opened this workshop by asking 17 B2B operators how many had posted that week. Two had. Her fix is a 3-doc setup (tone of voice, platform guidelines, Notion idea capture form) connected to a daily scheduled Claude task. Drop an idea the night before and wake up to 5 platform-native drafts ready for LinkedIn, Instagram, email, TikTok, and newsletter, all written in your voice. Victoria says once Claude knows how you write, about 90% of the work is done for you.

→ Read the full article

📚 Resources from Catalyst

If you want to go deeper on AI content systems and writing at scale:

• Catalyst's AI Content Writing System. The GPTs, audit frameworks, and SOPs Catalyst uses to ghostwrite founder content at scale.

Latest video

I build a Claude Cowork agent that runs your brand through ChatGPT, Perplexity, Gemini, and Claude simultaneously, since each engine produces a completely different ranked shortlist for the same buyer query. The agent identifies real buyer search prompts, uses browser automation to run them fresh on each platform, and compiles a blended leaderboard showing exactly where you show up and where you're invisible. Midway I pause to run live AEO audits of two audience volunteer websites. One is a kids' health education platform already ranking without much effort, the other is a banking consultancy that needs to fix schema and an llms.txt file before content can register.

→ Watch on YouTube