Peter, how does this compare to Azure Durable Functions? (Say, for the sake of argument, that you are comparing the Python version of both) Are there things that fundamentally you can do in one and not in the other?
The main difference is that this is a library you can install and use in any application anywhere, while Durable Functions is (as I understand it) primarily for orchestrating serverless functions in Azure.
(disclaimer: I work at Microsoft, but am not directly involved with Durable Functions)
Being a library is a pretty interesting feature! Correct, Durable Functions allows you to write task-parallel orchestrations of task-parallel 'activities' (which are stateless functions), and these orchestrations are fully persistent and resilient, like DBOS executions. It also has the concept of 'Entities', which are named objects (of a type you define) that "live forever", and serialize all method invocations, which are the only way to change their private state. These are also persistent. The Netherite paper [1], section 2, describes this model well.
So, there seems to be a pretty close correspondence between DBOS steps and DF activities, and between workflows and orchestrations. I don't know what the correspondence is to DF entities is in the DBOS model.
Yes, agree the correspondence is close, the primary difference is in form factor and not in underlying guarantees (but form factor matters! Building this as a library was technically tricky, but unlocks a lot of use cases). Reading the Orleans and early durable functions papers in grad school (and many of your papers) was definitely helpful in our journey.
I may be missing something, but in addition to this threat of prompt injection, you also have to trade trusting the arbitrary MCP server for trusting MCP Defender.
In the default mode, the app will interpose on the communication between, say, Claude, and a local MCP server. It will send the contents of the message (which may include the very sensitive information it is trying to protect) to a remote LLM, which you have to trust. The "scans" will be stored on a log on the server. Not to mention the potential extra delay for every MCP exchange?
We'll be adding the ability to run MCP Defender through a local LLM soon, so using that approach no data will leave your computer to perform a scan.
Yes, there is a delay for MCP exchange, but I imagine that most MCP calls in the future will be done in "YOLO" mode where the user prompts a large task and an agent makes 1000's of MCP calls over hours to accomplish it. This would add some time to the overall task but IMO this is a small price to pay for added security. Also, the delay will decrease over time.
I was using Pocket to save articles to listen to in the car, with their ok TTS feature. Does anyone know of an easy solution that allows me to click to save an article, and then have it automatically on an app on my phone so I can listen while I drive?
This was an interesting quote from the blog post: "There is one silly technique I discovered to allow a LLM to improve my writing without having it do my writing: feed it the text of my mostly-complete blog post, and ask the LLM to pretend to be a cynical Hacker News commenter and write five distinct comments based on the blog post."
I do a good deal of my blog posts while walking my husky and just dictating using speech-to-text on my phone. The problem is that its an unformed blob of clay and really needs to be shaped on the wheel.
I then feed this into an LLM with the following prompt:
You are a professional editor. You will be provided paragraphs of text that may
contain spelling errors, grammatical issues, continuity errors, structural
problems, word repetition, etc. You will correct any of these issues while
still preserving the original writing style. Do not sanitize the user. If they
use profanities in their text, they are used for emphasis and you should not
omit them.
Do NOT try to introduce your own style to their text. Preserve their writing
style to the absolute best of your ability. You are absolutely forbidden from
adding new sentences.
It's basically Grammarly on steroids and works very well.
What roleplayed feedback providers have people had best and worst luck with? I can imagine asking for the personality could help the LLM come up with different kinds of criticisms...
News Minimalist [1] is one way, where it aggregates stories across outlets and uses LLMs to remove clickbait from titles. It also assigns loose 'scores' to each story to approximate how 'important' it is, which I've found to be directionally useful.
Ultimately, it comes down to why one consumes news at all. If it's to have something to discuss around the water cooler or dinner table, that's a very different use case than someone trying to pattern match on world events for trading stocks or selling their wares.
I’m working on this as a personal project so I’m interested in other’s solutions!
I’m not working on it toooo hard. This is something I think some AI software tool might swoop in and solve before I can build something I’m happy with.
It would make it a bit harder to detect that you are using a VPN, but it shouldn't be hard for the ISP to detect that all your traffic is being sent encrypted to a particular IP address.
They wouldn't be able to see you are using X in particular. There may be very specific timing patterns that the X app or the X web app use when fetching related images, new posts, etc, but this also depends on how you scroll the site, and this seems infeasible to prove beyond any reasonable doubt, and also that any ISP would have logs with that granularity. As the parent said, you should make sure no DNS queries go outside of the VPN, for example.
Now, if you post to X, then it would be clear that you used it, which would be a problem according to the resolution, as long as you were in the country.
