Damir Vandic

Limux update 0.1.13 is out! If you were having trouble on x11, Arch, Fedora (RPM), or AVX-512, this one's for you! - zoom persistence - split pane ratio fixes - xdg data dir persistence fix - mouse tab focus fix - unix socket hardening - rpm/build reliability work - ghostty avx portability/test fixes download: github.com/am-will/limux

@simi9701 @dhh @Dell instant buy

@dhh @Dell please @Dell

In celebration of Omarchy 3.5 being the first distro to ship with complete Linux compatibility for the new XPS Panther Lake laptops, @Dell made me a special unit with super + omarchy keys instead of Windows and Copilot. So damn cool!

new herdr ux update: agent sidebar now toggles between "current" and "all" "current" shows agents in the active workspace. "all" shows every agent across all workspaces. small change, but when you're running parallel agents across projects, it saves a lot of context-switching.

Starting today, Plannotator is the best place to run and evaluate agent review findings against your code. I've been running CC and Codex reviews daily. Now you can too (you must be a paying user of each).

Going to be running a lot of side-by-side code review between Codex and Claude Code every day.

Due to popular demand, I am proud to announce my brand new course! Mastering Anthropic's Terms of Service: A 12-Module Masterclass "Finally understand what you're allowed to build." -------- Here's a quick peek into the course and what you can expect: Module 1: Is Claude Code a Product or a Suggestion? Module 2: The -p Flag: Personal, Professional, or Please Don't Sue Me Module 3: ACP: What It Means and Why Even Anthropic Isn't Sure Module 4: Can You Ship an App or Just Stare at One Lovingly? Module 5: Advanced Interpretive Reading of the 1,300-Page SDK EULA Module 6: Case Studies from Developers Who Tried and Disappeared Bonus Module: How to tag Anthropic employees on @X for Clarification ------ Join 43,000 confused developers. Lifetime access. No refunds. Link in comments 👇

I don't know what the fuss is about. Anthropic's rules on using subscriptions are very simple: Claude Code = OK Claude's online platform = OK Agent SDK running in personal software = OK... ish? Agent SDK running in commercial software = NOT OK Claude Code running in CI = ?? Oh, maybe it's not so simple... Agent SDK running in CI = ?? claude -p running in CI = ?? claude -p running in personal software = OK claude -p running on open source software, but run on my personal computer = ?? claude -p running on distributed sandboxes, kicked off by me = ?? Distributing open source software which relies on claude -p, and documenting how to use your subscription with it = ?? A thousand other edge cases = ?? Let me be clear. I have never before experienced, from any developer tool, such a frustrating lack of clarity over the basic terms of usage. I personally asked, 3 weeks ago, and have received nothing but delays. The recent @bcherny announcement did absolutely nothing to clarify things. I say this as someone who just released a Claude Code course - my incentives all align with supporting Anthropic.

I just shipped Swarm 0.1 for herding your agents, not terminals. Coding agents exhibit high latency because they make mistakes and get off track. However, running them in parallel can improve your development throughput by allowing you to tackle multiple problems at once. However, managing a bunch of parallel coding agents is a pain. Every agent needs a copy of the source code, a git branch, and multiple terminals. Before long, you're drowning in tabs with no idea what's running where. I built Swarm to fix that. It's a manager for all your repos, workspaces, and agent sessions, available as a native Linux desktop app powered by Ghostty. Here's how it works. You add a repository (your source code), create workspaces (git worktrees per task), and then open as many terminal sessions as you need in each workspace to run coding agents, development servers, and more. One workspace per feature. One agent per workspace. No more chaos. Works with any coding agent. 🔗 github.com/penberg/swarm

That indeed fixes the issue, but I still have some other issues, like the invisible popup when trying to exit or refresh, and I notice now something else: when starting a new session, when typing the first character a space is inserted first, the cursor seems not stable. But for now I can at least read the characters :), thanks!

I've patched and released "superterm up" to accept --disable-gpu on its start-up. After updating via "sudo superterm update", you should see the Chrome inspector showing: window.__supermuxTerminalDiagnostics() rendererStatus: "canvas" After this, I think we'll have to weight for @DHH to weigh in - perhaps open an issue on his repo? github.com/basecamp/omarc…

When we made superterm free for personal use, I recorded a quick video in my garden with airpods. 🫡 Your agents don't die with SSH 🫡 Unblock them from your phone 🫡 Track your goals + plans in session notes 🫡 Keep using Ghostty - every tmux session shows up

@alexellisuk @dhh I checked: proper GPU drivers are installed and working here. My Intel Arrow Lake / Arc Pro iGPU is using the `i915` kernel driver, and `vulkaninfo` succeeds with the Intel Mesa Vulkan driver (`Mesa 26.0.3`). Not sure if there's anything else I can check?

@dvic87 @dhh ( Having a look into your WebGL / GPU acceleration issue ) Does your system have proper GPU drivers installed? Here's "pi" with the default theme and a local model.

I tried Google Chrome and its an issue there, however, there I did see this error in the console: [.WebGL-0x1db40011c200] GL_INVALID_OPERATION: Error: 0x00000502, in ../../third_party/angle/src/libANGLE/renderer/gl/TextureGL.cpp, allocateMipmapLevelsForGeneration:1593. Internal error: 0x00000502: Unexpected driver error.

@dvic87 I have never touched Omarchy, but daily drive on Ubuntu Desktop LTS with Gnome and on MacOS Tahoe/Sequoia. Also have people on WSL and even XFCE. Maybe some of the themes/customisations @dhh and co encode into the distro are doing this? Can you try stock Google Chrome?

Seems like its triggered after starting Pi agent, then when I hit refresh the screen fades (I think there is pop up or something?) But after I hit enter, the rendering issue is fixed. And then when I start the conversation again the issue reappears. 2. Direct 3. No issues in Chrome inspector, I see in console xterm renderer: WebGL 4. No PiHole or ad-blocker

@dvic87 2. How are you serving the page? Direct or over some VPN/tunnel? 3. Then after that, open the Chrome inspector (or equivalent) - any console errors/404s or network issues? 4. Do you have PiHole or some ad-blocker extension running? No ads, but may be blocking CSS/JS etc.

@alexellisuk Omarchy (arch linux), default Chromium browser, its not related to the agent, its also in the terminal. I'll see if I can play around with the theme settings.

@dvic87 Hi Damir - what system are you running on? Browser, OS, Claude version? Not seen anything like this before 🤔 You can change font and theme via the command pallet. That might also help.

@alexellisuk Any idea how I can debug this ^^, font issue?

@alexellisuk Ahhh clear, didn't receive the email immediately but now I did, thanks!

Just shipped my first ever Pi Extension - Pi-listen Install: pi install npm:@codexstar/pi-listen Voice input for Pi CLI that works better than Claude code. (I was frustated using it) Hold space, talk, release. Words appear as you speak. 56 languages. Auto-detects sox/ffmpeg. Needs Deepgram Key on deepgram.com - (Get $200 Free - will run for a year if you don't yap too much) Pretty easy to set up. SoX recommended for better audio quality: brew install sox @badlogicgames @mitsuhiko @nicopreme Thank you for Pi and tools!

Tomorrow will roll out first set of clanker capabilities. It is a simple ask-ai interface. The foundation will enable better ai integration; more shift-left of clanker code review capabilities within your local context. Things like semantic diffs as well. Will look forward to feedback/ideas. Trying not to shove this stuff in anyone's face. The ability to ask Q's on the fly has been nice.

Always amazed by the people who dm me in support of the anti-Anthropic stuff. Feels like half the industry is scared to poke the bear. Thankful I am in a position to fight back on the bullshit. Will do my best. Sorry in advance if we get Claude banned on T3 Chat 🙃

T3 Code now supports Claude. If you have the Claude Code CLI installed and signed in, you can use it with T3 Code. Hopefully the lawyers won't make us remove this 🙃