Your AI agent rewrites title tags, meta descriptions, headings…
Then nobody checks if it helped.
I built SEO Notebook: a free, open source logbook for every SEO change your agent makes, with real before/after results from GSC + GA4.
Works with any agent (built-in MCP). Deploys to Cloudflare in minutes.
Repo below 👇
Just ask your agent to deploy to cloudflare and give them the link
github.com/evankrob/seo-note…
It’s no where close. They must be shelling out serious money for these tweets.
I was wrong
I've been saying for months that open source AI models are 6 months behind frontier
They caught up. GLM 5.2 is as good as Opus 4.8
This changes everything. If you run GLM 5.2 locally no government can take it away. You become sovereign
And even if you run through APIs, its a fraction of the cost
The battlefield is different now. If open source is as good as frontier, and people have cheaper alternatives, governments can't be as quick to regulate. It will destroy the frontier AI labs
All of this is such a massive win for the people
If you are not paying attention to local models yet, you are making a tremendous mistake
Guy made a great tool and at the right time.
Wildest thing about it, is there have been plenty of established players in the market (social media scheduling) already but Postiz still found a way to slide into the market.
This is job security.
I'm sure it will be as useful as broad match or "smart" campaigns.
A Google patent describes the process of Google building custom AI-generated landing pages for when people click on search results, instead of going to your website seroundtable.com/google-pate… hat tip @glenngabe @NarwhalJosh @BrandonLazovic
Like this concept so I built it. Fairwayodds.com let me know what you think.
Ev retweeted
We offered 5 people a Porsche 911 GT3 RS if they could get @WisprFlow to make a mistake
It's the fastest and most accurate AI voice dictation app that's 3x more accurate than ChatGPT, Claude, or Siri.
Today, we’re finally launching on Android. Download now: play.google.com/store/apps/d…
As a part of the launch, we’re giving away 6 months of Wispr Flow Pro for free.
Like, retweet and comment ‘Wispr Flow’ to get it. Enjoy.
— Written with Wispr Flow
Didn’t need AI to build expired domain pbns at scale.
But now point them at AWS buckets for unique regional related IPs.
My AI agent finds expired domains, scores them, builds entire websites, and deploys them to production. All while I sleep.
Here's the full pipeline 🧵
I wanted to build niche websites at scale. The bottleneck was never the content, it was finding good domains.
Expired domains with existing backlinks are gold. But every platform that tracks them either charges $200/month or hides the good data behind a UI with no API.
So I pointed my AI agent at one of these platforms and said: "Figure out how this works."
Step 1: API reverse-engineering
The agent opened the site in a headless browser, logged in, and intercepted every network request.
Within minutes it had mapped the entire hidden API:
→ Authentication flow (CSRF tokens, session cookies)
→ Search endpoints
→ Result polling mechanism
Then it wrote a standalone Python client. No browser needed anymore. Pure API calls.
Step 2: Automated domain hunting
I gave the agent a list of 30+ keywords across different niches.
Every night at 2 AM, it:
→ Searches all keywords in parallel
→ Deduplicates results across runs
→ Pulls backlink metrics for each domain
→ Cross-references with a second API for keyword difficulty and search volume
Step 3: Scoring algorithm
Not all expired domains are equal. The agent scores each one:
→ Backlink quality (referring domains, authority)
→ Keyword relevance (does the domain match the niche?)
→ Price filter (auto-reject anything above my budget)
→ Red flags (spammy history, penalized domains)
Top domains get flagged. I review 5-10 candidates instead of scrolling through hundreds.
Step 4: Domain purchase
Once I approve a domain, the agent:
→ Checks availability via registrar API
→ Purchases it (hard budget cap, no surprises)
→ Points DNS to Cloudflare automatically
→ Configures SSL, caching, security headers
Step 5: Site deployment
This is where it gets crazy.
The agent:
→ Generates a full site structure (pages, content, internal linking)
→ Uploads everything to the server via API
→ Adds mandatory security headers (CSP, HSTS, X-Frame-Options)
→ Deploys a sitemap
→ Submits the site to Google Search Console
→ Verifies ownership automatically
From "domain purchased" to "live site indexed by Google" zero manual steps.
Step 6: Monitoring
After deployment, a separate cron job:
→ Checks each site is still live
→ Monitors indexing status
→ Tracks keyword rankings over time
→ Alerts me if anything breaks
The result?
I went from manually building 1 site per week to deploying multiple sites per night.
The AI handles:
• Domain research
• Purchase
• DNS + SSL
• Content generation
• Deployment
• SEO submission
• Monitoring
I handle:
• Approving domain purchases
• Strategic decisions
• Coffee
The real insight:
Most people use AI to write blog posts faster.
I used it to build the entire infrastructure that finds, buys, builds, deploys, and monitors websites end to end.
The AI didn't just save me time. It made an entire business model viable that wasn't worth doing manually.
That's the difference between "using AI" and "building with AI."
About that time!
Feel free to vote for whoever you want for President but nobody loves your soil and water like this guy! The campaign has hit a ground swell, I think we'll reach triple digits this year #writein #roberts2020
This is not any elite hacking. Check your DNS entries for dead ones or you’re a target.
xitter.cf/er/status/185034025405…
Someone has managed to open a Shopify store via a subdomain on the official Kyle Jenner (@KylieJenner) cosmetics store - They're currently ranking fairly well for a bunch of new affiliate terms.
Interesting exploit - Probably very illegal 😅
I really didn't think this would still work @iannuttall but here we are... 2 of the first 5
Building in public:
Started a #nocode app that has been in the back of my mind for a long time.
Have made a few iterations of it but keep getting stumped at a certain features UX.
I’ve redone it at least 4 times.
Some serious paralysis by analysis going on. How do people plan out UX?
Should I just go with the fastest solution to get it live?
#indiehackers
Podcast – @Hokiesmash_ASD & @TalkinACCSports preview 2021 NC State Football with @ER allsportsdiscussion.com/2021…
Ev retweeted
Knowing why you got a bite can be the key to unlocking a winning pattern, and there are more reasons than you think. #bass #fishing #bassfishing bassblog.org/four-different-…
































