xur17

Yeah, 1password or keepass are both fine solutions for password managers in my mind.

I never found a comfortable way to ensure all of my accounts had more than one U2F key associated with them. I always wished there was a way to have 2 physical U2F keys, one on me, and by adding one to an account, both would work.

As is, I had to either:

- Keep both on me, and add both - I am at risk of losing both at the same time

- Keep one one me, one in a safe - I have to keep track of which device I've added to which service, and periodically take the backup one out of the safe and iterate through the "new" services

I was never satisfied with either approach, so I ended up with an OTP app with backups.

This is a spot where smaller orgs are getting a definite advantage (both cost and speed). If you're a company with 20 - 50 people, you can probably get away with Claude Teams instead of Claude Enterprise, and pay 10x less in ai costs.

Interesting to see, but I also think more and more usage is moving towards subscriptions, which isn't captured by this metric.

(both my personal and corporate use has done exactly this)

If you intend to use VulnHunter on Anthropic's first-party platforms (Claude API / Claude Code), we strongly recommend enrolling first via the verification portal.

Has anyone actually had success with this? I applied for my company several weeks ago, and never heard back.

Does that mean this one won't be open source?

Which limit? Weekly or 5 hour?

I've been using it with hermes and some coding (with opencode), and I am getting a LOT more than one feature out of it, but the work is spread throughout the week.

Looks like I have access to gpt-5.6-terra and luna. How does one decide between gpt-5.5 and gpt-5.6-terra? Pricing is similar, but it's hard to tell if it's better..

I personally use opencode so I can swap between models and try different options. I'd say I prefer claude (fable and opus 4.8) so far, but curious to see where gpt 5.6 lands.

For personal stuff, I've been pretty happy with chatgpt's $20 plan. I believe it has considerably higher limits than claude's $20 plan, and it's enough for the personal stuff I play with (hermes, and some small coding stuff). Also allows me to keep up to date on openai models.

And debit / credit cards are horrible for privacy (name and address info is sent along with payments).

Agreed. To me this feels like the perfect solution for websites and ai crawlers. Instead of having crawlers paying proxy services and captcha solvers, they can pay the website itself. As a web scraper, I'd happily pay the website provider to get access if it meant easy access to the content. Heck, as a human, I'd pay to avoid the dumb captchas.

Android does this fairly well for me - it will list out the options as they are spoken, and you can click on them directly to type the number. I believe (but haven't seen this recently) it will pre-render the options before they are read if it has knowledge of that phone tree.

Speed and scriptability

I feel like the general consensus on most of the recent layoffs were that they severance was pretty generous. If it seems to be working okay, why do we need to introduce additional inefficiencies to get something we already have?

Maybe, but I have heard the exact same experience from my Dad and others that have worked with unions. It unfortunately seems to attract people trying to take advantage of the system. Personally if I were offered similar jobs, one with a union, one without, I'd take the one without.

Gemini thinks my name is my brother in law's name, and despite explicitly telling it that's not my name + digging through the settings, it still amusingly calls me the wrong name.

Even in those cases I'd prefer to just be able to natively search the content that has been lazy loaded. I've run into more than one website where the search functionality they bound to control-f is horrible.

Any idea way "5x or 20x more usage" means?

Don't you still have to worry about big bills since bunny bills based on usage?

Thank you! Just realized someone else recommended this a hour ago as well, oops.

Thanks for building this!

Feature request: I'd love to be able to share a search results page with a friend. If you update the url to include the location or zip code I am searching for (ex: https://www.govauctions.app/feed/12345) this would be possible.

I am hunting for houses - I've used it as an assistant that catalogs them (but it has this stored in multiple places and will give me the wrong list a lot). I also had it read through my emails and create a doc with my upcoming trips.

I'd say it's right on the edge of being useful, but given the number of bugs, it's not really that practically useful. It's moreso a glimpse into the future.

I have it installed in a VM, and overall it's fairly useful, but very buggy. Right now I can send it a message asking it something, and it won't answer. I typically have to follow up 2 or 3 times before I get an actual response. This weirdly used to work fine.

More broadly, I'd say if you don't have something to retire to (kids or otherwise), don't.

I've found this to be the case as well. My typical workflow is:

1. Have the ai come up with an implementation plan based on my requirements

2. Iterate on the implementation plan / tweak as needed, and write it to a markdown file

3. Have it implement the above plan based on the markdown file.

On projects where we split up the task into well defined, smaller tickets, this works pretty well. For larger stuff that is less well defined, I do feel like it's less efficient, but to be fair, I am also less efficient when building this stuff myself. For both humans and robots, smaller, well defined tickets are better for both development and code review.

It even has a sprinkle of the silly "wow very technology!" aesthetic pandering that's typical of EVs.

I've never heard it referred to that way, but you nailed the description.

Back in 2019 I created a Google Spreadsheet titled "family debts" that allows my family (4 siblings and my parents) to record when we owe each other money, and periodically settle up. I later learned that I recreated Splitwise, but having something like this with trusted folks has been hugely useful. We have over thousand entries, and use it constantly for splitting gifts, buying something at the store for someone, etc.

Om Friday after Thanksgiving I spent half a day building a telegram bot that accepts an address and a list of Amazon links, and in turn orders the item (at a discount since it uses my Amazon credit card), and adds it to the above "family debts" spreadsheet.

I really like the idea of programmable, trusted lending like this, and feel like it could be extended to other groups that you implicitly trust.

Yeah, if the telegram and whatsapp spam I get is any reading, limiting by phone number is not sufficient.

If it's an all out race between the different AI providers, then it's logical for OpenAI to hire employees that are pre-trained rather than training up more internally.

A lot of airlines have started doing this by "gate checking" bags.