Chaitanya

[Chaitanya @cnijh]

Built a lightweight Mac utility app to have a floating chat window for my @openclaw bot.

This has become my main window to chat with my bot on Mac. It's open source, link below🦞

[Flying Lobster @flyinglobsterai]

Introducing Flying Lobster 🦞

Always-on-top chat for your @openclaw agents on macOS.

→ Global hotkey (works over fullscreen)
→ Multi-agent switching
→ Open source & free

One command to install:
npx flying-lobster install

[Chaitanya @cnijh]

@MatthewBerman

[Chaitanya @cnijh]

@MatthewBerman Got slightly better results from Fabel 5 using the Claude Design harness + had to prompt it to supply its own assets. Still prefer GPT 5.6 though

[Chaitanya @cnijh]

@thekitze https://t.co/9NVQFmxRhu

[david friedberg @friedberg]

if @elonmusk paid 100% of his net worth ($1.4 trillion) as a tax it would only cover federal government spending for 77 days. this isn’t a tax problem…

[Chaitanya @cnijh]

@alexanderrX_ I herniated L5 disc while lifting my toddler from her crib. She is 30 lbs.

[Chaitanya @cnijh]

@jacksondahl Re: Dystopian - How is any different than trying out a new piece of clothing in front of the mirror?

[Chaitanya @cnijh]

@_sholtodouglas @amorriscode Love the ChatGPT iOS app. That’s my go to on mobile. For anything serious, CC on terminal

[Chaitanya @cnijh]

@rajpals40432962 @laltvs2312 @officejigsaw @Drunkenpotatos1 @TapBit2769 @raymondtrader01 @mimi_sugawara @avalenz82 @tommy2times1212 @Rohit5118770480 @gerrudwallaert @Rakas88Lfc @o_sadeghian @lok8888888888 @rotilihsc4 @RamonaChandran @PlateaPR_ @MgMyatmin @sergejschets @reginaconr23 @trferni5 @nikitabier anything in the works to address spams like these?

[Chaitanya @cnijh]

@levelsio This has been my experience as well. I onboarded my dad and wife to openclaw as well. Now I’ve onboarded them onto Claude cowork because quality beats convenience

[Chaitanya @cnijh]

@kepano That would be awesome. I submitted an Obsidian plugin to import X bookmarks directly using the API. You need to pay X api pricing though, which is now is pretty reasonable.

[Chaitanya @cnijh]

@danieltian same! https://t.co/CulNw9ZhMj

[Chaitanya @cnijh]

Wow! This is a masterclass in product thinking and marketing. I just preordered. Haven’t been this excited about a new product since Tesla model S launch a decade ago.

[Chaitanya @cnijh]

Wow! This is a masterclass in product thinking and marketing. I just preordered. Haven’t been this excited about a new product since Tesla model S launch a decade ago.

[Aaron Tan @aaronistan]

We are hosting the largest showing of Lume in Palo Alto

Bring your own laundry

Details below

[Aaron Tan @aaronistan]

We are hosting the largest showing of Lume in Palo Alto

Bring your own laundry

Details below

[Chaitanya @cnijh]

@thekitze I wonder if GPT is optimized to keep the chat going (engagement farming) which leads to this behavior. Kinda like “I can show you 5 more ways to write this …” endless prompts on ChatGPT

[Chaitanya @cnijh]

@thekitze 💯

[kitze @thekitze]

my experience with gpt 5.4 in openclaw
- pls continue
- yes pls do that
- yes do what i told you to do
- yes continue
- please do what you said you're gonna do
- did you do that? no. then pls do it.

just pure pain

[Chaitanya @cnijh]

@rohildev Very cool. Happy to try it out if you’re looking for feedback

[Anything @anything]

BREAKING: Apple is scared of vibe coding

they removed Anything from the App Store so we moved app building to iMessage

good luck removing this one, Apple

[Manthan Gupta @manthanguptaa]

Love what Dhravya has been building and his work. He understands memory, so I am a little taken aback by this article. I am going to push back on this pretty strongly. This is being framed as a breakthrough when it's mostly benchmark engineering + system inflation.

First, 115k+ tokens are no longer a hard problem. We already have 1M+ context windows in production. Calling this "long-term memory" is stretching it.

Second, LongMemEval is no longer representative of real systems. It's been around for a while and doesn’t capture long-running agent workflows, real-time updates, noisy tool outputs, and cost/latency constraints.

Third, the "99% accuracy" claim is misleading. If you run 8 parallel prompts and count any correct answer as success, you are not improving memory. You are increasing the probability of a hit. That's not intelligence but sampling until something works. (spray and pray)
Same with the 12-agent voting setup. It's essentially majority voting across multiple attempts, which inflates benchmark scores without improving core capability.

Fourth, saying "no vector DB needed" is not an actual unlock. You have just replaced retrieval with multiple LLM passes, which require higher compute, greater latency, and higher cost. It is shifting complexity, not removing it. These things matter a lot in a production system, so you can forget about deploying this in production.

Fifth, the core claim that "agentic retrieval beats vector search" is oversimplified. The real problem in memory systems isn’t embeddings vs agents. Its relevance filtering, temporal consistency, memory lifecycle (what to store/forget), and grounding vs. hallucination.

None of these are solved here.

Also, this entire system assumes perfect extraction during ingestion, relies heavily on prompt engineering, and offers no guarantees of consistency across runs. So calling memory "solved" is very premature.

This is a good experiment in agent orchestration, not a memory solution.

[Aaron Levie @levie]

Having agents run over the weekend to do work for you will start to become entirely normal. There’s 0 reason why you wouldn’t add 2 extra days of output to your company with a couple prompts and a machine running in the background. Going to get wild.