@jack why does weebly suck so bad?
Not only is there no documentation of an MCP or other AI connection, there are no working dev docs at all:
https://t.co/EVcKqeBt8G 🤦♂️
Does anyone use this?
@outlandishjosh yeah, kind of a crapshoot.
I've had decent results with Claude creating powerpoints I then manually edit, but there was all the hype on the new image model.
Clearly my expectations were off.
Tried using gpt image to create a rate-card. I provided the company logo and website. 5 separate attempts with follow ups and it hallucinates a new logo every single time. I'm using pro, but it's still failing spectacularly.
@thdxr@apeatling Def true. The human needs a clear understanding of the tasks. The model needs solid context management, clear specs and scope in the prompt, and appropriate tools and skills available.
What else has a big impact on results?
This prompt is legit.
For @RepoPrompt users, I've written a bespoke multi-agent workflow you can install and use now.
Here's the code:
https://t.co/0VBAcaMTS1
We are investigating unauthorized access to GitHub’s internal repositories. While we currently have no evidence of impact to customer information stored outside of GitHub’s internal repositories (such as our customers’ enterprises, organizations, and repositories), we are closely monitoring our infrastructure for follow-on activity.
@emollick seems like a pessimistic take. Is it the "more vs less" language that bothers you?
Do non-coders equate "less-technical" with "not as competent"?
Yes, X's Paid Partnerships policy explicitly prohibits the label for "geo-political, political, social issues or crises for commercial purposes." Dinesh's post is political commentary, so displaying the paid partnership disclaimer would violate the rules unless X granted a case-by-case exception.