How to Build a Personal Liability Insurance Watchdog with RSS and AI

The Morning Scroll That Leads Nowhere

Every morning, I do the same thing: grab my phone and see what happened overnight. Dozens of unread messages in WeChat, a fresh round of trending hashtags on Weibo, a long push notification list from news apps, and then the tech forums—each one screaming 'just published' or 'major update.' After fifteen minutes of scrolling, I feel a strange emptiness: I know a lot of stuff, but I can't name a single thing that actually matters today.

It's not that I lack information. It's that there's too much of it. A new smartphone drops, an AI model updates, a company announces a restructuring—each item looks worth clicking, but the pile leaves me with almost nothing. We used to worry about not having enough information. Now the problem is the opposite: we're drowning in it, and being drowned isn't just a time-suck. It erodes your judgment about what deserves attention.

So I decided to run an experiment: build a personal news-filtering system using AI agents, taking back control of what I read every day. The same logic applies to liability insurance, where a single new ruling or regulatory shift can change how you assess risk. The trick isn't to read more—it's to read better.

Step One: Stop Scraping, Start Curating

My first attempt was simple: set a scheduled task in an AI agent to gather five tech news items every morning at 8 a.m. It worked, sort of. Instead of jumping between news sites, apps, and social media, I'd wake up, make coffee, and find a neat summary waiting for me.

But the novelty wore off in days. The same product launch, written three different ways, took up three slots in my briefing. Yesterday's news, rephrased by another outlet, would pop up as 'latest' again. The AI wasn't slacking—it was just pulling from a messy pile of 'bulk information': official announcements, media reports, secondhand interpretations, reposts, and clickbait. Give that to any AI and it'll summarize fast, but it can't judge what's genuinely worth your time.

The fix starts with the sources. Over the years, I'd saved 161 RSS feeds. But quantity doesn't equal quality. To stay sane in this info flood, you need a clear filtering and management strategy.

Action One: Build a Tiered Source Pool by Trust

In an age of recycled content and AI-generated fluff, the core principle is traceability. Information loses context and gets distorted with every retelling. So you want to get as close to the original source as possible.

For liability insurance, that means:

• Primary sources: official court rulings, regulatory filings (SEC, DOI), and insurance company press releases. These give you unvarnished facts.
• Authoritative media: Reuters, Bloomberg Law, Law360, and major business dailies. They have editorial standards and cross-checking.
• Secondary and aggregators: Insurance Journal, PropertyCasualty360, and niche newsletters. They excel at explaining complex rulings in plain English.
• Experts and analysts: Actuaries, risk managers, and law professors on LinkedIn or Twitter. They offer interpretation and context.

If you're feeling extra, subscribe to a few independent newsletters. They're often the best sources of all.

Action Two: Organize Your Feeds Like a Tree

When you have a hundred-plus feeds, you need a system. I use Folo, an RSS reader that lets me organize subscriptions into categories. It's old-school, sure, but nothing beats it for active aggregation and control.

I've split my feeds into six main buckets: technology, gaming, culture, AI, automotive, and now—liability insurance. Each category has its own subfolders: 'auto liability,' 'professional liability,' 'product liability,' 'cyber liability,' and so on. Opening Folo feels like flipping through a magazine I've personally curated.

Action Three: Make Your Agent Talk to Your Reader

Here's the magic part: Folo has a CLI tool that turns my RSS reader into a database my AI agent can query. Instead of random web scraping, the agent reads only from my pre-filtered feeds. And every item includes a direct link, so the AI never hallucinates a source.

I upgraded my morning briefing prompt to use Folo's API. The agent reads the last 24 hours of unread items and applies 'editor-in-chief judgment': merge duplicate coverage, cross-check details, and prioritize primary sources. If multiple outlets report the same court decision, I get one synthesized summary with a 'multi-source verified' tag.

Step Two: Train Your Assistant—It's Okay to Yell

Once the info pool is solid, a new problem appears: industry hot topics aren't the same as your personal interests. For a while, open-source AI models dominated tech news. Day one, I clicked. Day two, I saw a parameter breakdown. By day three, I knew it wouldn't affect my day. What I actually cared about were concrete hardware changes—a new laptop's specs, a phone's release date.

Humans can swipe past irrelevant news automatically. Agents can't. They just see that a topic is hot and feeds are still writing about it. So I told my agent directly: 'Too much AI news today. I want more consumer electronics and hardware. Remember this for future briefings.'

The agent created a MEMORY.md file and stored that preference as a long-term rule. And it worked—the next briefing was full of hardware news, with the AI models gone. That's what I love about agents: they don't read your mind, but they remember what you dislike. A good assistant is often trained by a bit of scolding.

For liability insurance, the same applies. Maybe you handle commercial auto claims and don't care about professional liability. Tell your agent. It will filter out the noise.

Step Three: Sew the Fragments into a 'Cyber Newspaper'

Even with good summaries, the chat interface felt flat. So I asked the agent to turn the day's picks into a clean HTML page—a 'cyber newspaper' just for me. The design was simple: minimal UI, card-based layout, with a title, core facts (multi-source), why it matters, and source buttons at the bottom.

The result was a clean, white-card layout that was a joy to read. Instead of a wall of text, I got a visual digest. Click a button, and you're at the original article.

But I took it further. I wanted to track a long-running story, like the rumored foldable iPhone. Rumors, denials, more rumors—each one seems like big news, but strung together, it's the same question being rehashed. So I challenged my agent to build a 'dynamic encyclopedia' for the foldable iPhone.

The prompt asked for a self-contained HTML file with: a 100-word status summary at the top, a tree diagram categorizing rumors by release date, form factor, screen and hinge, price, and unconfirmed specs; a timeline showing first appearance and current status; a keyword frequency chart counting independent sources, not reposts; clue cards with specific claims, credibility, status, earliest date, and original links; and filters by category, credibility, and status. The agent had to prioritize official sources, supply chain announcements, and analyst reports, and clearly label each claim as 'confirmed,' 'multi-source corroborated,' 'single rumor,' or 'unverifiable.' No unconfirmed content could be stated as fact.

The agent delivered a dark-themed, mobile-friendly page with SVG charts. The top summary was a no-nonsense paragraph on mass production status and key open questions. The tree diagram laid out all the fragmented info—from screen ratio to liquid metal hinges to price—at a glance. The timeline showed how each rumor evolved and where consensus formed. Each clue card had a credibility tag, and clicking the button took you straight to the original analyst report or Bloomberg article.

That experience made me less anxious about the foldable iPhone. Not because I lost interest—rumors can't replace the real thing—but because I could see which claims had solid sourcing and which were just clickbait. The fear of missing out vanished.

Why This Matters for Liability Insurance

You might be thinking, 'This is about tech news, not liability insurance.' But the method is exactly what you need for staying on top of liability insurance developments. Courts issue rulings, regulators propose rules, and insurers adjust policy language. These changes can affect your coverage, your claims, or your clients' risk profiles.

The same principles apply: rely on primary sources (court opinions, regulatory bulletins), cross-check with authoritative media, and use aggregation tools to filter the noise. Build a tiered source pool, organize your feeds, and let an AI agent synthesize the day's updates into a digest that's actually useful.

One thing I've learned: AI can collect, deduplicate, and organize, but it shouldn't outsource your judgment. The flood of information isn't going to stop. So instead of trying to swim faster, build a small dam upstream. Subscribe to sources you trust, keep a diversity of voices, and when you see a bold claim, go back to the original.

Whether it's a new smartphone rumor or a liability insurance ruling, the goal is the same: filter the noise, extract what's truly useful, and stay informed without being overwhelmed. That's a skill worth having.