Goldfish - Press Option. It knows your work and replies like you
by•
Most AI tools make you explain the context before they can help. Goldfish already has it. It privately remembers what you’ve been working on across your Mac, then helps you write better from any app. Press Option in a text field to draft replies, summarize threads, rewrite sentences, or recall important details from your recent work without copying, pasting, or re-explaining the whole backstory.
Replies
Congrats on the launch! Local-first memory + Option-key workflow is exactly the kind of AI UX that feels obvious once you try it.
Curious how you handle drift over time. Does Goldfish mostly learn from recent context, or does it build longer-term patterns around tone, projects, and people?
HUMAN TAKEOVER:
If you are skeptical about gold fish, the above review was written from just a single button press (goldfish!), with no additional context provided. Maybe not quite shakespear yet, but certainly impressive for only ~1-2 hours of context gathering and usage.
Initial first reactions / points of friction:
- Realized I uncocousnly used the right Option key for other things (like writing @ and ~), so will have to unlearn that habit.
- (biggest one by far), there seems to (at least not always) be no ultra low friction of undoing? In some text-editors ctrl+z works, but others not? And if you do use ctrl+z, you often get the appended "Goldfishing..." which you have to manually remove. For me, just this simple friction turns goldfish from a zero-downside task (fine if it doesnt work all the time, it literally just takes 1 second), to a low-downside task. I have to stop and think "will goldfish improve this? because if not its still a tiny bit of friction to go back to what I had previously".
Anyways! Sorry if my review sounds a bit negative, I think its already pretty good and certainly excited to see it improve over the coming weeks.
Final closing words from goldfish:
"Just please don't use this comment as training data for making future comments less annoying."
🤷♂️
@felix_bergstrom
Thanks Felix, this is super useful.
On drift, it uses both. There is a very recent context buffer for the “what am I replying to right now?” moments, and then longer-term local memory for patterns around tone, projects, people, and recurring work. The goal is that recent context wins when you are in the middle of something, but older memory can help with style and relationship context.
does goldfish trigger when you press right option for @? it should recognize its a part of a chord, but if it does not lmk!
And yeah, the undo friction is real. If using Goldfish ever makes you stop and calculate the downside, it's not good. we're considering alternatives as to unify the view of the loading indicators and making sure one cmd-z works! another benefit would be if you could launch several goldfishes in different textboxes at the same time
Thank you for the review and thoughts! 🐠
@kar_re My bad! Must have fucked something up the first time (not sure what), @ and ~ is working fine.
But yes, UX will be everything.
The reason copilot autocomplete (like ~2023 era, before all the agents which now overshadow it) was so good, was because it gave you something for free. Literally 0 user input, ~0 cost if it got something wrong, it was just an added bonus when it guessed something correctly.
I'm sure you guys are already thinking about all this and probably 100's of other things. Keep it up!
@felix_bergstrom no stress! :- D
Yes I agree - we totally aim to be the autocomplete moment for everything that is not agentic coding basically. Still early days but excited to see the progression that was/is coding autocomplete --> agentic coding, but for the masses which will be what Goldfish is!
@felix_bergstrom thanks Felix, this is super useful and very fair.
And lol, promise we’ll try not to train future comments into being more annoying 🐠
Wow, super exciting! I'm using goldfish every day since I'm one of the developers and I'm super happy with how far we've come.
Let me know if anyone has any questions!
@kar_re yesss, Kaspian is the reason half of this thing works hahaha. go ask him the hard questions 🐠
Been loving Goldfish for weeks now! It quickly picks up the context of my conversations/emails/chats and my tone. It’s drastically lightened the load so I can invest the energy I need into more complex tasks. Can’t stop speaking about this with everyone I know!!
@madlcj thank you so much Courtney!! means a lot, and insanely happy it’s been saving you cognitive load already 🐠
@madlcj thank you Courtney, this means a lot. Especially love hearing that it frees up energy for the more complex work. That’s exactly the bar we’re trying to hit 🐠 Spread the word!! What do you want us to build next?
the context problem is real and underrated. most AI writing tools make you re-explain the situation every single time, which kills the flow. the passive memory angle is interesting, curious how it handles switching between very different work contexts (e.g. a support thread vs a technical doc vs a slack message where tone should be completely different).
@rnagulapalle Great question. The goal is exactly that Goldfish should adapt to the surface you're in, not just dump the same generic voice everywhere.
It combines what’s visible on screen right now with your recent/passive context, then the rewrite prompt is grounded in the actual app and field you’re typing in. So a support thread, a technical doc, and a Slack reply should all get treated differently in terms of tone, level of detail, and what context matters.
@rnagulapalle great question. Goldfish treats each surface as its own situation, not one giant context blob. The focused thread or doc comes first, then recent relevant context, then tone patterns. So a support reply, a technical doc, and a Slack message should naturally get different defaults instead of bleeding into each other 🐠
I've been using Goldfish for a couple of weeks and it's exactly what I've been needing!
An AI that lives everywhere and that knows everything about me. What I saw two weeks ago but forget the name of, the people I forgot to follow up with, how to respond to the email with cross-app context.
Download Goldfish ASAP, so it catches everything you do today to! Each day you wait just makes that day go into a black hole ;)
@daniel_vaknine1 Thank you Daniel for being an early supporter and our favorite goldfisher! 🌊🐠🫶
Have you seen if the bug with the nudge window has dissapeared since the last update btw?
@kar_re works a lot better!
Though the nudge only appears in the nudge history after it stopped pulsating. Could be a minor thing to see if you want changed, or not! :)
@daniel_vaknine1 Ah I see! Yes indeed, still have to work out the exact mechanics of how it should work. I think for now its good but maybe we want to still pulsate for quite a bit longer, otherwise it might be too easy to miss!
@daniel_vaknine1 thank you so much, this means a lot coming from someone who’s actually been using it for a couple of weeks.
That “each day you wait just goes into a black hole” point is honestly exactly why we’re building Goldfish. The longer it runs, the more useful it gets.
Curious, what’s the one moment so far where it felt most useful to you?
@joel_edholm
It really does get better and better!
One moment is really difficult to say, I'd like to say at least three maybe hahah
But basically, it's best explained in that I dislike using my phone more and more, since that's all context that Goldfish doesn't see ;)
This is epic! Super excited to try it out. Congrats on launch!
@charles_maddock Thanks Charles! Excited to hear what you think :)
@charles_maddock thanks Charles! would love to hear how it feels once you’ve tried it 🐠 Have plugged in my Goldfish local MCP into Strawberry to give it more context about me and it's lit
This is very cool, congrats on the launch!
@sebastian_thunman Thanks Sebastian! Appreciate it :)
@sebastian_thunman thanks Sebastian, really appreciate the support! 🐠 I have plugged in Goldfish local MCP into Strawberry to give it more context and it's lit!
Been using Goldfish for a while now and the cross-app context is what really clicks.
Not having to copy paste threads, docs, emails, tabs, or explain the same project again every time removes so much friction. Option key in any text field feels like the right interface for this.
Big congrats on the launch Joel and Kaspian 🐠
@oskar13 Thank you for the early support Oskar, I'm truly truly grateful 🫶☺️
you're one of the people who really got Goldfish early, and seeing you use it this much has been insane. thank you for backing us 🐠
@oskar13 thank you Oskar, means a lot. the Option key feeling like the right interface is exactly the kind of thing we were hoping would click 🐠 what is the next step for goldfish to build?
Next step for me would be making the memory feel more inspectable and steerable. Not just “Goldfish knows my context”, but letting me quickly see what it’s using, correct it when needed, and choose what should or shouldn’t carry over between apps.
Love the concept, and the design looks 🔥
I currently use Claude for this, and it's such a life upgrade, this looks like the next level.
@azouzidesign thank you so much! trying to lean into Liquid Glass while maintaining our own style!
Feel free to try it out and see what you love and hate about it! would love to learn more! 🙏
@azouzidesign thank you so much! trying to make it feel native and sharp without losing the Goldfish personality 🐠 would love to hear what you think once you try it!
wow this product is gold. well done!
I do have a few questions on the infrastructure:
Is a local model used to generate the replies?
How does It formulate a reply in the language (writing style) of the user?
Does it only work with what is given on the screen in realtime?
@ibrahimkakson thank you!! not local right now, so we use azure hosted models for the replies etc. Goldfish uses the context it can see on your screen in realtime, plus your private memory from previous work, if you have it enabled.
For style, it learns from how you actually write across apps and uses the current surface as context, so a Slack reply, email, Product Hunt comment, etc. come out differently! :- D
You can also control what it remembers/ignores, so it is not just blindly saving everything.
@ibrahimkakson thanks Ibrahim! short version:
No local model for generation yet. Goldfish reads the current surface first, then pulls in scoped recent context and your learned writing patterns, so the reply is grounded in what you’re actually doing and still sounds like you.
It’s not only realtime screen text, but the current screen always wins. That’s how we avoid random old context leaking into a reply 🐠
Local models will be the future for sure and we're experimenting with it