It's not documented, but you can use skills.sh with Lovable! Here's how it works: 1. Go to skills.sh and grab the install command of one of the skills 2. Go to your lovable app and enter the prompt: "Run <command you copied>" 3. Lovable will install

Using the Credyt MCP server and Claude Desktop to visualize recent signups 😍

Love this event calendar Claude created to help me plan the upcoming Easter weekend with my daughter

Who is successfully using Claude as a scheduling assistant?

If you use GitHub (especially if you pay for it!!) consider doing this *immediately* Settings -> Privacy -> Disallow GitHub to train their models on your code. GitHub opted *everyone* into training. No matter if you pay for the service (like I do). WTH github.com/settings/copil…

Jeez Claude is slow (in GitHub actions) for me this morning

trying to run /schedule in Claude and getting "Unable to resolve organization UUID" ???

Love managing my Credyt account via the MCP 😍

🍿🍿🍿

Building got 100x easier in the past year. How to monetize? Well that's something else entirely.

I need to punish Claude

My attempt to create a basic PA using Claude failed at the first hurdle - it can't update project files/context - it can't push to GitHub - it can't send emails (only drafts) - it can't run things on a schedule (I need cowork) Is my only solution to build it in Claude Code

Set up Nous Research Hermes Agent as my personal assistant. With the help of Claude it took around 30 minutes to go from zero to chatting with "Aria" (my new PA) over Telegram.

So far using Nous Research Hermes with Qwen 3.5 Flash has meant I've had to train my PA on how to be PA. Almost every attempt to schedule something in my calendar has failed. So cost efficient, yes. Smart, no. What other cost efficient models should I consider?

That was inevitable. But wrapped up in a nice shit Sandwich from Claude (free credits)

My 2c after a few days running a Hermes agent for a few days is that if you're an engineer and have a good idea of what you want the agent to do, you'll still move faster building the orchestration yourself. I've yet to reap any benefit from the always on nature of the agent

Model test with Hermes agent was a bit of a fail tonight. Hermes 4 on OpenRouter didn't work because it lacks tool call support. Gemma 4 was slow AF and then just started timing out.. So I went back to Qwen 3.5 Flash. My dumb PA is better than no PA. Sometimes.