irrationalfab

Agreed. It is ironic that in the AI race, the real differentiation may not come from how smart the model is, but from who builds the best application layer on top of it. And that application layer is built with the same kind of software these models are supposed to commoditize.

Interesting! Generating PDFs with properly paginated content is still a pain point in 2026.

Do you have a comprehensive integration test suite that can validate the robustness of your implementation?

Prince starts at 2k. This is OSS

Handy is genuinely great and it supports Parakeet V3. It’s starting to change how I "type" on my computer.

Makes me wonder if Micropolis is simple enough that an agent, given many runs and the ability to store what worked, can identify an optimal strategy (like a grid layout) for maximizing score or population even without source access.

+1... like with a large enough engineering team, this is ultimately a guardrails problem, which in my experience with agentic coding it’s very solvable, at least in certain domains.

Agent/MCP/Skills might be "Netscape-y" in the sense that today's formats will evolve fast. But Netscape still mattered: it lost the market, not the ideas. The patterns survived (JavaScript, cookies, SSL/TLS, progressive rendering) and became best practices we take for granted.

The durable pattern here isn't a specific file format. It's on-demand capability discovery: a small index with concise metadata so the model can find what's available, then pull details only when needed. That's a real improvement over tool calling and MCP's "preload all tools up front" approach, and it mirrors how humans work. Even as models bake more know-how into their weights, novel capabilities will always be created faster than retraining cycles. And even if context becomes unlimited, preloading everything up front remains wasteful when most of it is irrelevant to the task at hand.

So even if "Skills" gets replaced, discoverability and progressive disclosure likely survive.

There's a pattern I keep seeing: LLMs used to replace things we already know how to do deterministically. Parsing a known HTML structure, transforming a table, running a financial simulation. It works, but it's like using a helicopter to cross the street: expensive, slow, and not guaranteed to land exactly where you intended.

The real opportunity with Agent Skills isn't just packaging prompts. It's providing a mechanism that enables a clean split: LLM as the control plane (planning, choosing tools, handling ambiguous steps) and code or sub-agents as the data/execution plane (fetching, parsing, transforming, simulating, or executing NL steps in a separate context).

This requires well-defined input/output contracts and a composition model. I opened a discussion on whether Agent Skills should support this kind of composability:

https://github.com/agentskills/agentskills/issues/11

Anthropic is positioning Claude in Chrome as a beta feature for day to day use, expanding beyond the earlier research preview.

- Pull data from dashboards into one analysis doc

- Address slide comments automatically

- Build with Claude Code, test in Chrome

https://www.youtube.com/watch?v=rBJnWMD0Pho

> If those features aren't supported by the widget's hard-coded schema, you're out of luck as a user.

Chat paired to the pre-built and on-demand widgets address this limitation.

For example, in the keynote demo, they showed how the chat interface lets you perform advanced filtering that pulls together information from multiple sources, like filtering only Zillow housers near a dog park.

This feels like the death of the app, and the rise of the micro-app.

Very interesting tech. Ephemeral, high-fidelity preview environments that require zero setup are a key enabler. They let you rapidly validate changes within the complete context of a web or mobile app, accelerating feedback loops and cutting friction for minor updates. This also empowers business users to safely implement small, self-contained UI adjustments which is particularly powerful when combined with LLM-driven suggestions.

This nails a real problem. Non-trivial PRs need two passes: first grok the entrypoints and touched files to grasp the conceptual change and review order, then dive into each block of changes with context.

Like convex.dev

Ironically, LLMs might make it very hard for new frameworks to gain popularity since they are trained on the popular ones.

Thank you so much for this!

I consider that talk one of the most interesting about programming and I consider it underrated as well.

Some context: Appie Durán (https://twitter.com/AppieDuran) is the dog of Eloy Durán (https://twitter.com/alloy), the original author of CocoaPods.

Credit goes to Florian R. Hanke (https://twitter.com/hanke) for exploiting this weakness of sans-serif fonts.

The Silver Searcher is very good for using the terminal as an IDE.

I have two aliases which I find very helpful.

- Search Obj-C files:

  alias ago='ag -G ".*\.(h|m)"'

- Search Ruby files:

  alias agr='ag -G ".*.rb"'

Also if you are looking for an alternative to sed don't miss replace[1].

[1] https://github.com/harthur/replace

Have you seen https://github.com/glock45/iOS-Hierarchy-Viewer?

Interesting twist. Excellent execution. Great MVP.

What business model would sustain the cots?

Why isn't this open source?

Thank you for making this. I like it and I will give it a try, nonetheless the first impression is good.

As the other comments point out the copy is very bad. I think that you should improve it and that HN commenters should check things out before writing. To be honest I've only tried because of the backslash (I don't like to install extensions from non authoritative sources).

Finally, this is a nice extension... but if a frame is your concept of ambition, you should reread PG's essay.

---

This is an extension that replaces google with a sidebar loading immediately the first page that you search. It allows to browse among the search results quickly.