Haven't built anything on the agent skills platform yet, but it's pretty cool imo.
On Android the sandbox loads an index.html into a WebView, with standardized string I/O to the harness via some window properties. You can even return a rendered HTML page.
Definitely hacked together, but feels like an indication of what an edge compute agentic sandbox might look like in future.
It's not that the context window is adequate, but rather an agentic LLM can search the source of truth using appropriate tools (SQL, term search, etc.)
RAG made sense when the semantic search was based on human input and happening as a workflow step before populating context. Now it happens inside the agentic loop and the LLM already implicitly has the semantics of the user input.
Living here the decline is tangible. And this is West Oxfordshire; not one of the poorer parts of the country.
An example in microcosm: a local village suffered road flooding due to failed maintenance of water pipes. Our rent-seeking privatized water company effected the minimum repair required by regulation.
The next section of old pipe burst almost immediately, flooding the road further for most of January, utterly destroying the surface, through the road base in many places. Even at a crawl it's difficult to avoid tyre damage.
Over a month later the water repairs were effected. Then shortly after some local roadwork notification signs were put up.
Those expecting repairs to the moonscaped road were disappointed: instead the relentless bureaucracy of British local government installed traffic calming measures on top of the broken road, as the work had already been booked and could not be stopped by any means as even basic roadworks lack any degree of dynamism in their execution.
All this still needs to be made right. These small scale failures will compound and compound until the entire state is drowned in the consequence of its incompetence.
We need to recognise the difference between the GP rant and what you're describing. The austerity is undeniably still reverberating through the country. It will take years for this ship to turn around, although it is being turned around. For example, in just about a month we're getting European-style rents with the Renter's Rights Act, which is transformational. We can and should do better, and everyone can contribute to solving those issues, but after a decade of nothing the necessary changes are finally being implemented.
But the rant is entirely counterfactual. Britain is a very rich country with beautiful and recovering nature, a healthy and educated population, one of the more capable armies in Europe, a functioning deterrent, and a relatively healthy political system. We just got two new parties becoming credible threats to the "main" two (regardless of the parties' views, the political competition itself is a much healthier situation than the American duopoly)! We just abolished hereditary peers, which is a constitutional change (and it can just be done)! Below the everyday media noise, we're doing alright as a democracy.
The UK is still a respected "brand" in most of the world despite what chronically online people say. British education is the most sought-after in many countries for example.
It's important to realise that the US is full of fascists obsessed with the perceived decline of Europe. They love to shit on Europe. I think it's about distracting themselves from the abject moral, political and economic failure of voting for Trump twice.
> An example in microcosm: a local village suffered road flooding due to failed maintenance of water pipes
Your example only compares against the UK past.
It has zero relevancy because it says nothing about relative change against other countries.
Anecdotally for the USA, I went to New Orleans last year, and I was stunned at the rotting infrastructure. Coming from New Zealand, the USA seems to be trying to copy the trajectory of Argentina.
Then again, I see serious problems in my hometown (e.g. sewage treatment plant) and country (e.g. big problems with rail, ferry, air, electricity, 3 waters). Apart from the societal issues that it seems all countries are facing.
I know what a rebuilt city looks like, because I come from one. Hurricane Katrina was 2005. Christchurch Earthquake was 2011. In my opinion, my home town has recovered better and faster from destruction than New Orleans has.
I also live within a floodzone. There is a high probability I will learn how we deal with flooding in the future (different flooding - shallower and lacking the winds and hopefully better pre-planning for avoiding harm).
> everything looked brand new
Absolutely not, to me.
And the conversation is regarding infrastructure. A bunch of Christchurch infrastructure is brand new.
Fyi, this is not true. California has them but they are not routine, and are a function of internal political dysfunction that is quite unique to California. The grid here is still extremely fragile, and vulnerable to e.g. cyberattack and other disasters, but let's not get carried away.
It helps with cost, agreed. But caching doesn't fix the other two problems.
1) Models get worse at reasoning as context fills up, cached or not. right?
2) Usage expansion problem still holds. Cheaper context means teams connect more services, not fewer. You cache 50K tokens of schemas today, then it's 200K tomorrow because you can "afford" it now. The bloat scales with the budget...
Caching makes MCP more viable. It doesn't make loading 43 tool definitions for a task that uses two of them a good architecture.
Claude will load the name and description of each enabled skill into context at startup[0]; the LLM needs to know what it can invoke, after all. It's negligible for a few skills, but a hundred skills will likely have some impact, e.g. deemphasizing other skills by adding noise.
You could be thinking of the Claude tool search tool [0][1], which lets it do that on-demand loading. Skills (like another comment mentioned) have a progressive disclosure mechanism built-in. Agent Skills [2] appear to be the common format across most of the tools, but I have no idea if this actually has tool provider buy-in.
Having seen a family member die absolutely horrifically in a matter of weeks due to late-diagnosed pancreatic cancer, I'd consider suicide if I got the same diagnosis.
On Android the sandbox loads an index.html into a WebView, with standardized string I/O to the harness via some window properties. You can even return a rendered HTML page.
Definitely hacked together, but feels like an indication of what an edge compute agentic sandbox might look like in future.