I have a client (a used car dealership in the US) who wants me to build an AI-powered "Lead Hunter Agent" that would: - Scan TikTok, Instagram, YouTube, and Facebook using Apify or similar tools - Detect posts, comments, and videos related to buying used cars - Identify users showing buying intent - Automatically engage with them (comments, DMs, outreach) - Capture and store those leads in a CRM I've pushed back. Here's why: - Geography: The dealer is local. Social media users could be anywhere in the US. No reliable way to filter by location when scraping. - No contact info: You only get usernames. No email, no phone. The only way to reach them is on the platform itself. - Automated engagement = banned: Meta has been mass-banning accounts for bot activity since mid-2025. You can't automate comments or DMs in Facebook groups without getting flagged. - Local Facebook groups (like "Atlanta Used Cars") do solve the geography problem — but you still can't automate conversations there. That's a human job, not an AI agent. - Cost per lead: Apify credits + LLM tokens + dev time + extremely low conversion rate = likely more expensive per qualified lead than just running geo-targeted ads. My recommendation was: skip this, invest in geo-targeted ads and local SEO. Explore Facebook groups manually first before automating anything. The client ran my analysis through ChatGPT and Claude asking them to "roast" it. Got confident counterarguments. Then when he prompted "assume the specialist is right" — both AIs immediately agreed with me. So: am I wrong? Has anyone actually built a working automated social media lead gen system for a LOCAL business? Looking for real experience, not theory. Thanks.

Hey builders, Quick question for the community. Many of the workflows I’m running with Claude Code generate documents in .md format, which is great for structure and automation. The problem comes when converting them to PDF… they work, but they’re missing that super professional look (clean layout, typography, spacing, visual hierarchy, etc.). What I’m trying to find is a Claude Skill or workflow that can take .md files and produce really polished, client-ready PDFs. Ideally something that: - Applies professional layouts automatically - Handles titles, sections, callouts, tables, and spacing nicely - Produces PDFs you could send directly to a client or use in a report Does anyone here know a skill, tool, or workflow for this? Even better if you’ve personally tested it. Would love to hear what you’re using 🙌

Hello everyone, lately I saw a really cool marketing agency on IG named mobileesitingclub. I was wondering can connect Claude with my Higgsfield so that Claude can write prompts in the same style ? Does anyone have experience with that? BR Naha

GPT 5.4 dropped last week, and it felt like the right time to talk about something I've been thinking about: how much more we can actually trust AI agents to finish a task now, especially with the newest models. The real power of these models right now isn't in a chat window. It's in setting them up inside your terminal or an IDE, where they get actual freedom to do things they couldn't do before (file access, running code, hitting APIs). That's where the magic happens. MCP server hype has died down, but I think it is a great time to talk about them again. If you don’t know what that is, it is basically a custom server an AI agent can search to access all of the API endpoints of specific software. Here's the thing, you don't really need one handed to you anymore. Writing custom instructions for how to use a specific software platform's API lets you basically build your own. You're just treating each platform like a tool the agent can reach for when it needs to. Now, safeguards still matter. You probably don't want your agent taking out a $5 million loan on your behalf (clearly). But there's a huge range of tasks where it can get to 90, maybe even 95% done without you touching anything. And that last 5-10%? That's where you step in for the finishing touches instead of doing the whole thing from scratch. TLDR: I am realizing I can use claude code for more shit.

So the question is n8n versus make.com? Which one is better,? They both have their ups and they both have their downs. I’m looking for some personal and professional opinions of how you feel about each program. N8N in my opinion, builds agents better and easier and I feel can do more. Make.com seems to be easier to be working with all the way across but more difficult in building out agents. I like using both platforms and both have their strong points and both have their weaknesses. The other thing that I look at with them is their cost. Self hosting with NAN is great. Don’t get me wrong. It is awesome. I do find a bit of a nuance with the self hosting and that it is something else in addition that I have to keep up with. make.com on the other hand is not self hosted, but is cloud and again building agents are a little more difficult n8n and cost wise make.com is not bad. The other thing is with NAN doing the cloud posting versus self hosting $25 a month for 2500 actual workflow usage is phenomenal. You’re not being charged for each node that is being used in a workflow. But how many times the work flows are triggered. Meaning you can have one workflow and it can run 2500 times for the $25 a month. The big question is what is everybody else’s take an opinion on this matter? I know everybody prefers make over N8N or N8N over make just looking for valued opinions and discussion