wonderfuly

Just self-host on Cloudflare: https://github.com/wong2/cf-mailroom

Interestingly, although this is about hardware, it contains the protocol for communicating with Muse, so I packaged them into a separate TypeScript SDK, https://github.com/wong2/muse-client

The simplest use case: you can chat with Muse in your favorite IM.

This feature has just been added!

What file type do you need?

Why is it more reliable? The human usage point is fair, but I doubt how long it is still necesary.

Why cli instead of just HTTP API doc? The agent can use curl or write code to send requests.

It is account based, and I'm using Clerk for the auth.

Last year, I came across NotebookLM and immediately noticed a pain point: importing the web pages I was browsing into NotebookLM required several steps. So in less than a day, I developed this Chrome extension: NotebookLM Web Importer[1], which allows for one-click importing. As NotebookLM has gained popularity this year, my extension has also seen great growth. So, in July, I added paid premium features to unlock additional features. It exceeded my expectations and quickly went over $500 a month. It now has over 100,000 users and is still growing.

[1] https://notebooklm-web-importer.com

I'm a victim of this.

In addition to concerns about npm, I'm now hesitant to use the GitHub CLI, which stores a highly privileged OAuth token in plain text in the HOME directory. After the attacker accesses it, they can do almost anything on behalf of me, for example, they turned many of my private repos to public.

A curated list of Claude Skills: https://mcpservers.org/claude-skills

A curated list of Claude Skills: https://mcpservers.org/claude-skills

I'm building https://notebooklm-web-importer.com a Chrome extension that enhances NotebookLM UX

I'm doing the same everyday, but with https://chathub.gg

This article tries to argue that the AI bubble has burst by pointing to the failed release of GPT-5. Admittedly, the release of GPT-5 was somewhat of a flop, but I think it's more of a failure in its launch rather than the model itself. In fact, if you use the GPT-5 Thinking model, it's actually quite good. They attempted to make the model automatically route to different levels of thinking intensity, but the routing didn't work very well, which led to the various bad cases people experienced.

Chat now: https://app.chathub.gg/chat/cloud-gpt-5

Here is my projects: https://bento.me/wong2

I'm waiting for you to open remote engineering positions

Hope to be added in.

Available here: https://app.chathub.gg/chat/cloud-llama4

You can see the replay of the AI agent working on a task here: https://manus.im/share/OZB4PvsXY5N5FrvLXqqCl4?replay=1

You can also use https://chathub.gg/repo2txt

ChatHub is the first service to do this, and it's been around for almost two years, even before the release of the GPT-3.5 API.

Chat now: https://app.chathub.gg/chat/cloud-gemini-2.0-flash

You can also use https://chathub.gg/repo2txt

Chat now: https://app.chathub.gg/chat/cloud-qwq-32b

Where did you get the data?

Awesome, I once developed a project called https://rerss.xyz, aimed at creating an RSS feed that reorders historical blog posts, but it was hindered by the two issues mentioned in the article.