RC

OpenClaw - Perplexity Edition

"We used to debate using tabs vs spaces in code we'd type out"

Claude Code watching me write code manually

You can now connect to Claude in Chrome using the VS Code extension. Use it to debug frontend apps, collect data or automate your browser. Install the extension and type @ browser to get started.

Codex skill for Claude because why not!

I'm Boris and I created Claude Code. Lots of people have asked how I use Claude Code, so I wanted to show off my setup a bit. My setup might be surprisingly vanilla! Claude Code works great out of the box, so I personally don't customize it much. There is no one correct way to use Claude Code: we intentionally build it in a way that you can use it, customize it, and hack it however you like. Each person on the Claude Code team uses it very differently. So, here goes.

Claude Opus 4.5 (left) is running faster than Sonnet 4.5 (right)

This opens up Claude Code, the "TUI" to be usable with a lot of models out there! 🙌

This is rare but valuable comparison table from them..

Anthropic is donating the Model Context Protocol (MCP) to the @linuxfoundation and establishing the Agentic AI Foundation alongside Google, OpenAI, Cloudflare, and others! Long live open source!

Wow so now I can stream my fav Game of Thrones episodes on Netflix without having to worry about what app to install: Max Or HBO or HBO Max. bloomberg.com/news/articles/…

I personally believe Claude Code (specifically the terminal CLI written in Node.js) as the best developer product of 2025. But daily use with Anthropic models gets expensive fast - $15/M output tokens adds up. Was exploring open-weight models, found that DeepSeek v3.2 works out of the box with Claude Code. They built an Anthropic-compatible API endpoint specifically for this. Tool calling works. MCP works. $0.42/M output - that's 35x cheaper. Setup is just env vars: I made the above shell functions to switch between them. use-deepseek for exploration and bulk work, use-claude (unset vars) when I need the original experience. I would rate the performance close to Sonnet-4.0 in terms of getting things done. Open Models are getting serious. Not sure if @openrouter supports `/anthropic` compatible APIs, that will be even more convenient

We get the missing manual now..

Woah, crazy how much better Gemini-3 is than other flagship models

You think you've seen it all, FSD actually has It's trained on what amounts to over 100 years of real-world driving scenarios from our fleet of >6M vehicles A lifetime of driving scenarios in 10 minutes

Generative AI can respond in seconds. But speed means nothing without trust. @_summation delivers both — connecting every metric and system into a single source of truth so enterprises can instantly analyze performance, surface verified insights, and yes, run executive-level QBRs before you arrive at the office. 👉Schedule a demo summation.com/request-demo?u…

This is great!

f I read another post with “literally” and “AI” together on my feed, I am literally going to ask an AI to generate a filter that deletes them all!