Viewing profile — gavinuhma

Awesome feedback. I’ll think through this. The sandbox is spot on; Control what the server can do. Especially important when running locally

I’ve made the mistake of typing “--tools echo” before instead of “--tools=echo”, just in case that was your error too.

Thanks!! Hrm, re that error: What does “god --version” say? The log might not show up until you get a successful connection. I’ll look into that. Thanks for trying it out!

Hey everyone, I've wanted an easy way to control which mcp server tools are available to clients. So for example, I might want a gmail server to only expose the read tool (but not …

Good question! Uploaded documents get converted to embeddings within the Nitro Enclave (NE), and then the embeddings are encrypted with a key that only the NE has access to. When t…

You can use CapeChat UI to mess around with it: https://chat.capeprivacy.com/ Or you can also create a free API key here: https://app.capeprivacy.com/api-keys to use the interactiv…

I redacted the full wikipedia paragraph with the API. Like, the nickname "The Great One" is a pretty major tell! [NAME_GIVEN_1] [NAME_FAMILY_1] CC ([NAME_GIVEN_2] [NAME_FAMILY_2]; …

Exactly. A super famous person like Wayne Gretzky is really hard to protect. For fun, you can try to tease ChatGPT with information like. "Who is [Name1]?", it won't know, but then…

Yikes, we'll have to remove that. It's a really old course on privacy-preserving machine learning from 4 years ago and has nothing to do with this product despite the generic name.…

Entirely local and 0 sub-processors is the ideal! I hope we are trending that way as an industry

Good points. I think the rabbit hole of OpenAI sub-processors is not commonly understood. The humans at TaskUS are moderating prompts, and then you have Azure, CloudFlare, and Snow…

Good question, some developers implement a manual approval step, so you can review the redacted prompt before you submit it rather than making it automatic. It depends on their pro…

Yep! The more you can do locally the better. An entirely local LLM is the best for data privacy and security. Any time data leaves it poses some risk. The de-identification itself …

That's right. So in the case of credit card numbers we redact it like [CREDIT_CARD_NUMBER_1], [CREDIT_CARD_NUMBER_2], etc so the LLM can still answer prompts like "how many", but i…

It's a great question. Redaction limits the LLMs ability to draw on the underlying training data on the subject. This can work to the developers benefit in many cases, like asking …

We’ve built the Cape API so developers can keep sensitive data private while prompting LLMs like GPT-4 and GPT 3.5 Turbo. With Cape, you can easily de-identify sensitive data befor…

"there is little consistency either in the definition of ultra-processed foods or in examples of foods within this category" source: https://www.sciencedirect.com/science/article/p…

This is super cool! Where does the data come from?

Will this work with docusaurus?

That’s exactly why this is needed

Nice! Thanks for sharing. I also learned about shellcheck thanks to this thread, which has been super useful

Pr for that https://github.com/gavinuhma/checksum.sh/pull/4 I ended up trying with process substitution so no tmp file. It works. Trying to decide if it’s more difficult to read

Agree. Although checksums are smaller and easier to copy/paste. Same with a url. I download the script from A, and the checksum from B. And then I verify them locally. So A and B b…

Haha gold

Ok should be fixed now. Appreciate you pointing it out. That whitespace trick is really interesting