One of the most original ideas I have seen on HackerNews in the past few years.

tldr; devcontainers let you completely containerize your development environment. You can run them on Linux natively, or you can run them on rented computers (there are some providers, such as GitHub Codespaces) or you can also run them in a VM (which is what you will be stuck with on a Mac anyways - but reportedly performance is still great).

All CLI dev tools (including things like Neovim) work out of the box, but also many/most GUI IDEs support working with devcontainers (in this case, the GUI is usually not containerized, or at least does not live in the same container. Although on Linux you can do that also with Flatpak. And for instance GitHub Codespaces runs a VsCode fully in the browser for you which is another way to sandbox it on both ends).

Basically we do not rationally analyze what work can be automated and what work is forever safe. We just assume that "sexy work" is safe, and work backwards to figure out how to explain this belief to ourselves.

Actual accessibility is a different topic. Honestly I don't know much about the a11y of CLI apps in general. Is there something that makes Taskwarrior bad in that regard? Is it about the way it forms things like list outputs or tables?

for instance here's how you'd download all the unread blog posts by combining blogtato and wget:

blog .unread read | wget \ --recursive --level=1 \ --page-requisites \ --convert-links \ --adjust-extension \ --no-parent \ --directory-prefix="$HOME/offline-posts" \ --input-file=-

For instance, `wget` is a pretty widely used HTTP client that is able to mirror links for offline access. Then you can use standard tools such as `grep` to search in all the offline content. And `blogtato` does already have an export feature, so it is almost trivial to write a script that saves all posts for online content.

So perhaps what should be done here is find a user friendly tool for offline access/search for web content and just add some convenience features to `blogtato` so that they integrate very easily.

Points: 1

# Comments: 2

Bug Tux Racer vibes btw! I think adding some music and perhaps snowfall could make it even more immersive

edit: and i think you could easily make it feel more detailed by adding a texture for the snow,and increasing the resolution of nearby terrain by modulating it with some precomputed fractalized perlin noise. So basically "free" detail that only exists in the rendering logic, does not influence the complexity of the heightmap. and perhaps you could replace the trees with sprites that are images of real trees (stylized to the visual style you're going for)

Then again, the success of such a project might depend on other factors. Given the complexity of the task, I can imagine that just "lucking into" the right design decisions early on could have a major impact.

It seems like it's fully community-maintained, there is no big company or foundation behind it. Honestly it's hard to believe!

There was just one major problem, the infamous "topological naming problem" which caused issues downstream is you edited a non-leaf node. That was pretty frustrating to deal with, but in later releases they fixed it I think. (Have not tried it since because I didn't have anything to model)

honestly, that in itself would be valuable, though simple.

This idea, i. e. actually "moving" context form one agent to another is even more interesting.

Model Context Shell lets AI agents compose MCP tool calls using something similar to Unix shell scripting. Instead of the agent making each tool call individually (loading all intermediate data into context), it can express a workflow as a pipeline that executes server-side.

Since the orchestration is deterministic and reproducible, you can also use it with Skills.

Tool orchestration runs outside the agent and LLM context, so the agent can extract only the relevant parts of data and load those into context. This means you can save tokens, but also you can work with data that is too big to load into context, and your agent can trigger a very large number of tool calls if needed.

Also, this is not just a tool that runs bash - it has its own execution engine. So no need for full system access.

Example query: "List all Pokemon over 50 kg that have the chlorophyll ability"

Instead of 7+ separate tool calls loading all Pokemon data into context, the agent builds a single pipeline that:

1. Fetches the ability data 2. Extracts Pokemon URLs 3. Fetched each Pokemon's details (7 tool calls) 4. Filters by weight and formats the results

At least in it's current iteration, it's packaged as an MCP server itself. So you can use it with any agent. I made this, and some other design choices, so you can try it right away.

Comments URL: https://news.ycombinator.com/item?id=47073185

Points: 3

# Comments: 0

Here's a tip: GitHub now allows you to embed a proper video in your README. (https://stackoverflow.com/questions/4279611/how-to-embed-a-v...). Quality would be much better, and people can navigate back-and-forth in the video.