One of my fav things about @goose_oss (GLM-4.7 @Zai_org ) is it's ability to access previous chats.
Had a chat crash and wouldn't reload.
Opened a new chat and had it grab context from the previous chat and continued on.
It was also able to fix the crashing chat via sqlite😍
When you tell your coding agent to implement api security...
It goes ahead and as a test creates a random api secret ... without displaying it
Then sets production to it and locks you out of your own system. :-) @augmentcode #itworks
1
37
This is exactly how I feel when building with LLMs. I have no desire to write much code anymore but I love building things, automating and solving problems.
Web development is fun again ✨
ma.ttias.be/web-development-…
3
57
Just built a Chrome Bookmarks MCP using @goose_oss and @Zai_org GLM-4.7
Using it to go digging through a lot of digital packrattery.
Source code is at github.com/RobAMills/bookmar…
Inspired by github.com/mixophrygian/brow… which is very useful too.
Rob Mills retweeted
sora2 is making this easier than ever
this ai health page is doing $94k/month just by reposting videos
$41k from youtube
$33k from instagram
$20k from tiktok
no filming
no editing
no team
just ai content, repurposed daily across platforms
anyone can hit $10–20k/month using the same model in almost any niche
rt + comment “healthy” and i’ll dm you the method
(must follow for dm)
Rob Mills retweeted
Pinterest + ChatGPT + Ideogram = $19,516
And I can prove it works in next 5 sec.
I'm running 6 Pinterest blogs that generates ~$4K each.
So I'm giving away my ENTIRE system for free:
• Complete Traffic Guide
• 10 Premium Pinterest Templates
• Like & RT
• Comment 'send'
Rob Mills retweeted
This Pinterest Blog earned $1800 last month and you can work toward the same Results.
Grab my📌60 Premium Multi-Niche Pin Templates📌 packed with proven, high-performing Pin Designs you can fully customize for your own blog.
1. Follow + RT this post 👇
2. Comment PINNER and I’ll DM you the Gumroad link.
I just made a Chrome extension with @cursor_ai that will block sites until I unblock them.
I'm trying to resist the urge to add more features cause it's so fun.
Source: github.com/RobAMills/JustFoc…
#vibecoding #buildinpublic
3
78
Rob Mills retweeted
We put together for you a list of the most viral content on Facebook in Q4 2024, including the EXACT Pages for you to follow and the exact posts for you to study 🤯
Scrape these pages, analyze these posts, create your own versions of them and go off and make more money than ever before from your Facebook assets 👇
It's the perfect starting point for anybody looking for exact pages to follow and exact content to share all the way down to the actual links to the post themselves.
And, of course, we're giving this document away for FREE!
Retweet + reply "MONEY PRINTER" and I will send it to you.
Must follow for DM 👍
If anyone is looking for domains, @sean is a trustworthy source.
This quoted post is unavailable.
2
360
Rob Mills retweeted
These viral math puzzles work great for Facebook engagement! 📈
I've created 1,000s of unique ones. Do you want a FREE bundle of 25 to try out on your social media pages?
You got it!
✅Comment "MATH" + Retweet
✅Must Follow (So I can DM you the link)
Whoa! When did Bulk Domain Search come back on expireddomains.net? It's one of my favourite features that was gone for a while. #domains
114
Rob Mills retweeted
Built myself a Python automated internal linking algorithm. Want the script? Get it below.
Been having some auto-gen content fun lately, with a maximum focus on making stuff as natural as possible.
Interlinking is now my focus for it.
Built my own interlinker last year, but well, WordPress is a pain with custom editors and I gave up.
Looping back to it now, and I have the base for Python interlinking script that will link my clusters together. Prioritise links how I want them.
It will take the Markdown formatted text, pre-Wordpress upload, and sort internal links for me.
Add keywords, get links.
Natural anchors
A few will be exact match links, but most will have natural surrounding text that forms a (mostly) coherent anchor.
Keyword: Blog
Anchors:
a blog
blog posts
bloggers
a blogging setup
this blog content
all possible blog posts
Keyword: WordPress
Anchors:
all those lovely WordPress features
WordPress
all the WordPress bloat
We're looking at 2,000-2,500 word articles, so I'll be looking at inserting 3-4 links to other articles in the cluster, and then a link to the pillar for the cluster.
Fallback link forcer
But, not all articles in the cluster will have a keyword that will get it links.
In that case, I've got a little fallback that will force a link in for any URLs that don't match a link.
Just one of those old "read more:" type setups, but this has gotta be better than nothing.
What it does now
Now, this is just the basis of it for testing. This isn't a run, get links, setup. It's the algorithm for it. The brains.
It still needs to be tweaked to dynamically list the keywords/urls, and also to actually pick which link to insert, as it's just giving you all the options right now.
I'll be having it randomly select a link, and then blocking it from selecting any other link in the nearby sentences, but you can obviously tweak this how you'd want.
Built with ChatGPT.
You can modify it with ChatGPT.
ChatGPT can even help you set python up... It helped me!
You could;
- Block exact match
- Block phrase match
- Add synonym matching
- Add multiple links per URL
The list goes on!
Want the script?
Like, repost and chuck a comment down below and I'll slide into your DMs with the script.
Geezuz @seanmarkey Is a link from WashingtonPost and BusinessInsider really worth this much these days?
1
94
Some Monday morning #pseo inspiration. elevation.maplogs.com , sunrise.maplogs.com lat/long and sunrise/sunset times. Ahrefs says it ranks for 326K keywords 😯. Weird that there's no homepage in the root directory of the domain tying the two sub-domains together.
1
84
In addition to what @Seanmarkey says here I find that a domain with a high DR but ranking for a low # of keywords will usually be because the domain has been dead/dormant for a while. The Wayback Machine will show limited or no crawls after a certain point.
NameJet and SnapNames April monthly report: SEO Edition - analyzed by @seanmarkey
domainnamewire.com/2023/05/1…
How am I supposed to get any "real" work done during the day after reading this? Thanks for sharing @iannuttall
SEO traffic > display ads is a shaky business model 😰
It's a good starting point, but the gravy train doesn't roll forever.
Here is a pSEO + AI flow that I've been using recently with great success:
3
295