Saumitra

Clawdbot safe use: docker or vm or aws tier? What’s your pick?

I have a Mac mini and I feel compelled to install ClawdBot!

@heygurisingh Claude

Claude can make you more money than anything before. So I built the Advanced Claude Mastery Course • 80+ Chapters • 1000+ New AI tools • 2000+ New ClaudePrompts 100% FREE for the next 24hrs only Just: • Like • Follow • Reply "Claude" I'll DM you a link.

everyone’s sleeping on gemini 3 pro for AI automation while some experts are quietly building 47+ n8n workflows in just 2 weeks using master prompts here's what most don’t know: gemini 3 pro has a 1m token context window it's built specifically for agentic tasks and tool use but raw prompting wastes 80% of that capability master prompts change everything: → feed it your full tech stack context → define node architecture patterns → specify importable json output result: production-ready workflows from plain english descriptions Want the full master prompts experts are using? 1. Retweet & Like 2. Follow me 3. Comment “GEMINI” i’ll send you the entire step-by-step prompts library for free, plus real-world use cases (must be following so I can dm you)

@dashboardlim Gemini

#gemini should add a way to invoke gems by adding them as @ gem

@elonmusk @grok not a surprise! :)

Grok

@itsolelehmann This is gold. Saved for later. Quick tip: claude skill generation now offers direct “copy to skills” button instead of needing a download.

@aakashgupta Have you compared make.com? I tried make and it worked quite well for me so haven’t tried n8n. During debugging corner cases, I wished I had a local agent instead.

I've been telling people n8n is the tool to learn for 2 years now. The company went from $270M to $2.5B valuation in 7 months. Almost 10x. Here's why I think most people are missing what makes this so interesting. Zapier charges per task. Every action in your workflow counts separately. A 5-step automation costs 5 tasks per run. n8n charges per execution. The entire workflow counts as one. A 200-step AI agent pipeline costs the same as sending a single Slack message. The math compounds FAST. Let's say you're running a competitor monitoring workflow like the one Pawel built in this episode. It pulls from Google Sheets, hits Perplexity 6 times, compresses context with code, sends to GPT, converts markdown to HTML, and emails you. On Zapier that's maybe 15-20 tasks per run. At $70/month for 2,000 tasks, you can run that workflow 100 times before you're out. On n8n you get 2,500 executions for $20/month. Same workflow runs 2,500 times. The delta widens with complexity. And AI workflows are ALL complex. What's driving n8n's growth isn't just the pricing arbitrage though. It's that 75% of their customers are now using AI features. Revenue grew 10x in one year. User base 6x. The timing is perfect. Every company needs to automate AI workflows. Most companies don't have engineers to spare on building custom pipelines. But Zapier's task-based pricing makes complex AI automations prohibitively expensive. n8n sits right in the middle. More powerful than no-code tools. More accessible than writing code. And the pricing model doesn't penalize you for building sophisticated agents. NVIDIA's venture arm invested in their Series C. That tells you evrything about where they think the value is heading. The hidden insight from this podcast... Pawel showed three levels of building the same competitor monitoring workflow. Standard workflow with preset steps. Agentic version with some flexibility. Fully autonomous agent. The standard version used 5,000 tokens. The autonomous agent used 90,000 tokens. Same output quality. 18x the cost. Most people will see an agent like Claude or GPT working autonomously and think that's the goal. The people building production systems know the real skill is figuring out exactly how much agency you actually need. Every decision point you can remove from the AI saves tokens. Every output format you can define saves inference time. Every tool description you optimize prevents hallucinations. This is where I think the automation market splits. Zapier owns the simple trigger-action space. They're betting on non-technical users who want to connect two apps without thinking about it. n8n is betting on the messy middle. Technical enough to customize. Visual enough to iterate quickly. Priced for complexity. The multi-agent research system Pawel demonstrated at the end is the future state. Orchestrator agents distributing tasks to sub-agents. Each sub-agent searching, fetching, compressing context. Final synthesis by a copywriter agent. You cannot build that economically on task-based pricing. I don't think most PMs realize how quickly this is becoming a core skill. Understanding when to use deterministic workflows versus agentic ones. Knowing where to compress context. Recognizing which tool descriptions make agents succeed or fail. n8n just became the best classroom for learning all of it.

Product skills of the future: - Intuition about what's worth building - Clarity in describing the solution - Taste in knowing when it's great - Agency to do the above without being asked

@minchoi Good summary. How are skills shared in teleported sessions?

This is literally Claude Code Setup Cheatsheet Based on Boris, the creator of Claude Code. Save this.

Check out my latest article: How To Build Your First Claude Skill: Complete Guide linkedin.com/posts/saumitra… via @LinkedIn

Claude web skill builder now allows copying generated skill with one click! No need to download the zip file for enabling custom Claude skills on web. #claudeskils

Installed #opencode with #grok-code while waiting for claude and gemini tokens. I miss using my skills but this setup looks great for one shot trying some ideas. Maybe I could write something to interpret existing Claude skills?

It’s a Saturday morning 9:30AM. I’ve already ran out of #gemini and #claude tokens :|

also, we can’t ask async questions to Gemini? If app is switched, it stops any processing? Reminds me of Google collab sessions

Why is Gemini loosing my chats across devices?

One time, only one device showed the latest chats and others were stuck at old ones

@elonmusk Waiting for grok cli! 🚀

Major upgrade to Grok Code coming next month. It will one-shot many complex coding tasks.

@grok voice mode just chuckled? Interesting Easter egg! :)