Connor Adams

@mitsuhiko I would love to scroll up and still keep the text input box at the bottom - so you can read & answer questions (i.e. alternate screen) OpenCode and CLAUDE_CODE_NO_FLICKER=1 do this. I assume this is a fairly big change so I'm curious to hear any thoughts or recommendations.

If something really bothers you about pi, give us feedback please :)

@levelsio In the meantime you can ask your agent "We're getting DDoSed-find the attacking IPs and block them" with github.com/cloudflare/mcp as shown by @threepointone

The main thing I miss on Cloudflare is a single [ Block IP ] button It boggles my mind it's not there, you can see a traffic spike and a person doing 50,000 requests per minute, but you have to write a security rule and go through 10 steps to block it Too difficult Just let me [ Block IP ] from the Security Analytics or general Analytics dashboard @daneknecht @eastdakota

@NathanFlurry > tmux isn't cutting it for agents anymore what are your pain points exactly? ps. shameless plug for if you want to continue your tmux sessions on your phone - try 📱 remobi github.com/connorads/remo…

tmux isn't cutting it for agents anymore I love Conductor, but doesn't work over SSH nor sandboxes Looking for recs for any of: - tmux ext for agents - Sandbox-based solution that's tmux/nvim friendly - Hooks for CC/Codex/Amp/OC to give me a "command center" or "inbox" w tmux

@t_blom @conductor_build In the meantime, there's 📱 remobi github.com/connorads/remo…

We need a mobile app for @conductor_build

@bentlegen If you'd like to access your tmux session from mobile you might like 📱 remobi github.com/connorads/remo…

Also, made it so everyone can chat w/ the agents over Slack github.com/modem-dev/baud…

Running coding agents on a separate Linux box: 1) Connect remotely via Tailscale 2) Persistent terminals with tmux 3) Linux so everything's agent-scriptable 3) No personal data, less anxiety 4) Agents get great internet no matter where I am 5) Battery lasts forever Won't go back

@levelsio If you're into tmux you could try remobi - makes it mobile and touch friendly

Are you guys aware I am coding mostly on my phone now all day via Termius to Claude Code on my server while I go with gf to the dentist, clothing store, cafe, etc. 😛✌️

Interactive post 🔵 have a play 👇 connoradams.co.uk/blog/wtf-is-ai…

WTF is AI actually doing? 🤔 Forming your own mental model of how AI models work helps you use them well I like to think of them as four ingredients 🎱💬🧠🛠️, and this mental model helps me anywhere from prompting ChatGPT to setting up Claude Code for agentic engineering

@conar_app L'histoire de ma vie quoi

We probably need to start transitioning to the new name...

@NathanFlurry @opencode @rivet_dev @sandbox This is fun! Do you think it'd be possible to get the OpenCode web UI working with Gigacode over tailscale? So I can Claude Code on my phone, via OpenCode web UI, whilst I'm away from my computer. I've been trying to set it up but struggling - any tips.

Introducing Gigacode: Use @opencode's UI with any coding agent 🌎 Supports Claude Code, Codex, & Amp harness 🍽️ Not a fork (opencode attach) 🛜 Try remotely via HTTP 🛠️ Powered by Sandbox Agent SDK (@rivet_dev) Try it: npm i -g @sandbox-agent/gigacode (more options below)

@steveruizok You might like what @tokengobbler and friends have done with review.fast Example review walkthrough and narrative: api.vibekanban.com/review/2a032e2…

I'd like see if the agent could create a walkthrough changes inside of the IDE, a kind of "optimal loom" on top of the diffs. "Right, so the purpose of this PR was to implement such and such. In this first commit I'm updating the types for the new etc etc..."

The best code review tool I've come up with is asking Claude to reimplement the PR on a new branch in a narratively optimized perfect git history

@JetSetVideo @NicolasZu github.com/vercel-labs/ag… has a lot of next stuff in there

@NicolasZu Can you do it for NextJS too?

🛠️11 React Three Fiber (R3F) skills for Claude Code and Codex: animation, shaders, physics, postprocessing, and more. 📂 Bookmark it! Incredibly useful to build 3D games or 3D websites with LLMs contribute: github.com/EnzeD/r3f-skil… install: $ npx add-skill EnzeD/r3f-skills

@orkuhq @Michal_Krsik @opencode It would be great if they can get rid of that 128k limit. But looks like it might be around for a while at least. github.com/microsoft/vsco… Hopefully I'm wrong 🙏

@Michal_Krsik @opencode from my observations yes you can see it clearly when using models via opencode with the sidebar token/% windows usage

OpenCode can now officially be used with your Github Copilot subscription with the $39 pro+ subscription you get access to the best coding models wonderful to see them support open source and user choice of tooling in this way

@kitlangton This might be the best website I've ever used. Thanks, looking forward to the rest of the chapters 💕

It is time → effect.institute. 2026, the year of 𝚎𝚏𝚏𝚎𝚌𝚝. I'm extremely proud of how this turned out. There are many chapters to add, but I do hope you enjoy what's there so far. Let me know what you think. 🫡

@thdxr docs.exa.ai/reference/exa-…

does anyone know of a public mcp server with no auth?

@nummanthinks @aigirlagent @code Can you see this working on @opencode? What's the gap?

@aigirlagent @code It needs to be through the Codex CLI It has the best auto compaction for any coding agent right now

Today I ran two complex tasks through Codex with GPT 5.2 Extra High The first ran for 2 hours 30 minutes The second ran for 1 hours 45 minutes Both resulted in: - all acceptance criteria resolved - all test coverage complete - zero broken or non-working code Amazing

@thdxr #L49-L126" target="_blank" rel="nofollow noopener">github.com/connorads/dotf…

@thdxr After pressing always approve I often move them to my global config to avoid doing it again My general thinking is allowing things that are probably safe like reading files (in current project) and viewing git and running tests But I want to asked when adding new packages

those of you who use coding agents with permissions turned on for bash commands do you ever use "always approve"? doesn't it feel clunky to be approving some and not others on every single session start? tell me your mindset on this

@thdxr @johnsangwonsuh Jim come get your damn cluster

@johnsangwonsuh it's free!

this is the code it takes to deploy an elixir app with auto-scaling, auto-clustering, rolling deploys, and load balancing to aws

@AkitaOnRails Thanks for linking to my repo but here's the repo you're probably looking for, unless you love my guide for some reason 😅 github.com/jdx/mise