Activity
Mon
Wed
Fri
Sun
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
What is this?
Less
More
Memberships
Maker Zero: Claude Code, AI
791 members • Free
Maker School: AI Automation
2k members • $184/m
Claude Code Profit Room
388 members • $77/month
Oleg's AI Lab
7.1k members • Free
Easy Machine AI
2.8k members • Free
Automatable Free
18.1k members • Free
Chase AI Community
70.6k members • Free
AI Automation Club
7.6k members • Free
AI Automation Society
419.1k members • Free
21 contributions to AI Automation Society
4h •
Day 3 Xreview Skill
🔥 Day 3 of the AIS 7 Day Challenge Used Nate's Skill Builder for Claude Code to build a chatbot that logs my daily, weekly and monthly execution reviews from The Less Paradox (a transformational seminar by Lakers Kokmaiya at the Pillars Institute). Optimisation from today: split the planning questions and review questions into two separate groups — way cleaner flow, and the bot stopped mixing forward-looking prompts with reflection prompts. Biggest takeaway so far: the Skill Builder makes it stupidly fast to turn a personal framework into a working tool. On to Day 4 💪
0 likes • 2m
@Leon Yeap This is a great use of the Skill Builder, turning a personal framework into a working tool is one of the best things to build first because you'll actually use it every day. And splitting the planning questions from the review questions was the right instinct, forward-looking and reflection are two different mental modes and mixing them muddies both. One thing that takes a review bot to the next level: have it read back your last entry before it asks this week's questions. That way each review builds on the last one instead of being an isolated snapshot, and you start seeing patterns over time. That's where the real value of logging reviews shows up. Nice work on day 3.
9h •
#AISOS AI OS configured, loom video for showcasing one of its skills.
So this is the most interesting project I have done so far for sure... and I am already imgining multiple case scenarios in which this might be very useful for me, especially with complex workflows involving external databases, API's, and other stuff like that. One of the skills in this AI OS is to simply give me a rundown of the unread e-mails I had from the day I am requesting that skill to be used (i.e. the last 24 hours), and I can also create a to-do list on ClickUp for some e-mails that need urgent responses and rank based on priority. Very fun project, and this AI OS may have a huge amount of potential to be better. Loom video: https://www.loom.com/share/976fcdd8290e40a4bd2a87a27664e64c
0 likes • 2m
This is a great first skill to put on your AI OS, email triage is one of those things that quietly eats an hour a day, so automating it pays off immediately. The ClickUp handoff with priority ranking is a nice touch. Two upgrades when you want to push it: give the priority ranking a clear rule to follow instead of leaving it to judgment, like flag anything from a known contact that asks a direct question or mentions a deadline. Makes the ranking consistent run to run. And for the urgent ones, have it draft the reply too, so instead of just a to-do you get a head start you only need to approve. Turns it from "here's what to deal with" into "here's most of it done." Really solid build.
3h •
Day 6
#AIS Challenge I have created a Scheduled Task to check Lowest Flight Ticket Prices until next run From Toronto to Melbourne on Every Sunday at 3:00 PM on Weekly Basis. For Loop, I set a reminder to drink water and look away from screen for 10 seconds, and it will remind me after every 10 minutes. Also created one time reminder to take screenshot after 60 seconds for demo. I found Scheduled task more helpful, because It will let me schedule my skills to run by its own that I otherwise run manually. Even where I used for Day 6, is gonna be helpful to track ticket prices for a week. For loop, If You start learning new things, you can use loop to remind you something which you could forget otherwise. The Self-healing is a thing that I always praise these AI Agents. It makes the learning AI a piece of cake for People who think learning AI is difficult. Tools Used: Claude Code Pro, VS Code, Firecrawl
0 likes • 6m
@Lovedeep Singh Love that you're leaning on the self-healing, that's the feature that quietly makes everything else reliable. Most people build automations that break the first time an input changes, and self-healing is what turns a brittle script into something you can actually trust to run on its own. One thing that pairs well with your scheduled flight tracker: have it only ping you when the price drops below a threshold you set, instead of every run. Turns it from noise into a signal you'll actually act on. Solid day 6.
0 likes • 7m
@Edifico Re Investment Depends a lot on what you're using it for, but a few things that help across the board with the newer models: give it more context than feels necessary, it uses it well. Let it plan before it executes on anything multi-step instead of one-shotting. And be specific about the output you want, format and all, rather than leaving it open. The models are strong enough now that the bottleneck is usually how you frame the task, not the model. What are you mainly using Fable 5 for? Easier to give a useful tip with the use case.
2h •
Noobs vs pro vs Hacker ?
How do you differ someone who’s new in the ai space, against an intermediate and a pro. What are they doing differently at each level? Let me know you’re thoughts below, I am looking to level up!
0 likes • 8m
@Karl Stenor Great question. The way I see the three levels: Beginner obsesses over the model and the prompt. Which model is best, what magic words to use, chasing every new tool. They think the AI is the skill. Intermediate can actually build a working system. They've learned the tools and can ship a workflow that runs. Their focus is making it work. Pro focuses on everything around the AI. Reliability, cost, and knowing when NOT to use AI at all. They've learned the model is the least important part, the value is in the guardrails, the validation, and the judgment about what to automate. They also ship fast and get feedback instead of researching forever. Shortcut to level up: stop asking "which model" and start asking "where is my system likely to break, and how do I make that not matter." That question alone moves you up a tier.
1-10 of 21
Online now
Joined Apr 23, 2025
Powered by