@vimtor
Account based inSpain
About this account
- Account based in
- Spain
- Connected via
- Spain App Store
Account-level information from X, not a live location or the device used for a specific post.
it’s beautiful to do something right @opencode @anomalyco
Spain
Joined October 2017
- Tweets2.4K
- Following687
- Followers1.9K
- Likes7.7K
Pinned Tweet
i've joined @anomalyco to maintain sst and support @opencode
sst has always felt like the right abstraction - making complex but reliable infrastructure easy to use
it's an honor to work on that with people i've admired for years
nice to see
@opencode always ahead
and @vimtor always on top of new raycast changes
github.com/raycast/extension…
i've finally landed on using opencode and @mymind MCP
a nice UI to look at and a good "database" that organizes itself
i'm thinking of trying one of those agent memory providers like @supermemory for this
for this use case, i care a lot more about the information being accessible than about how the UI looks
i'm thinking of trying one of those agent memory providers like @supermemory for this
for this use case, i care a lot more about the information being accessible than about how the UI looks
any app that unifies watch later lists for movies, books, TV shows, and games?
honestly i don't even care that much about the UI. an agent-only interface is fine for this one
implementing the opencode provider NOW
This week we shipped an update that lets extensions bring their own AI models to Raycast 💫
You can then use those models in AI Chat, Quick AI and AI Commands.
Available today for Mac and Windows.
Full changelog 👉 raycast.com/changelog/
i don't know if it's because the models have gotten better or i have AI psychosis but lately i feel like i've been able to do months' worth of work in days
am i alone in this?
GPT-6 Luna just blew my mind
using Python, it was able to chain tool calls to search the internet and find this painting by the french painter and founder of impressionism Claude monet
guys you won't believe what GPT-6 Sol just did
using javascript it was able to search the internet and find this painting by the Dutch post-impressionist painter Vincent van Gogh
i've been testing opus 5.5 for a couple of years now
it's very good. it feels like an improvement over previous models like haiku, especially for coding tasks like centering a div
thank god for open models
i had deepseek scan for vulnerabilities so fable could validate and patch them
i was getting blocked when asking fable directly
i've made a very simple opencode plugin for quickly referencing a paragraph from the conversation
sometimes it's faster than voice prompting, especially when the agent comes back with a list of steps or ideas
install it in V2 by running:
𝚘𝚙𝚎𝚗𝚌𝚘𝚍𝚎 𝚙𝚕𝚞𝚐𝚒𝚗 𝚊𝚍𝚍 𝚘𝚙𝚎𝚗𝚌𝚘𝚍𝚎-𝚚𝚞𝚒𝚌𝚔-𝚚𝚞𝚘𝚝𝚎
learn more here: npmjs.com/package/opencode-q…
setting up a personal assistant feels a lot like configuring vim
you spend more time configuring it than you actually save
especially today when so many useful services don't have a native connection
