Alfred Naayem

My dear front-end developers (and anyone who’s interested in the future of interfaces): I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept): Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow

The missing feature for working with Claude Code on markdown in VS Code Annotate text right there in the rendered preview for Claude to see: Cmd+Shift+V to open preview H - highlight for discussion C - leave a comment for Claude E - request an edit D - mark for deletion plain-text syntax in the .md file. no sidecar files, no special format. works with Claude/Codex or any LLM that reads md. The vision for this is something closer to google docs suggest mode but right in VS Code for your AI to read. Find it in VS Code extensions → "Ace AI Markdown Feedback" github.com/41fred/ace-mar…

@thesamparr @Grantfw Have a lead would be down to talk about how we can partner on @Grantfw

My buddy is @Grantfw. His company is Rosedale. I originally invested because he started the company based on a dumb idea I had. Thankfully pitvotted last year to his idea and it’s booming. rosedale.ai

My buddy runs a company helping manufactures implement ai. He showed me the leads he’s getting. It’s nuts. Family businesses I’ve never heard of making $100m a year. They know the need ai but no idea what to do. Crazy how much momey is out there

Repo here: github.com/41fred/mcp-orp…

If you're running claude code with MCP servers (cloudflare, heroku, github, etc), you're probably losing CPU to zombie processes right now. When a session crashes, force-quits, or the terminal gets closed, MCP servers survive as orphans. each one pegging a core at ~50%, doing nothing. we found 11 running at once. it's been a known bug since june 2025. Issue: github.com/anthropics/cla… Our short term fix: a lightweight process monitor that detects orphaned MCP processes every 10 minutes and kills them automatically. no dependencies, just bash + launchd. until a real fix lands upstream. git clone github.com/41fred/mcp-orp… && cd mcp-orphan-monitor && bash install.sh

@bcherny we make a lot of automated meta improvements to items in .claude/. Last updated really broke this for us: v2.1.77 (March 17) — introduced a change where .claude/ is now a protected directory that requires explicit permission prompts even in bypassPermissions mode.

I built a Slack bot that runs Claude Code against your actual codebase. Not a generic chatbot. Your CLAUDE md, your repo, your existing Claude plan. No API costs. Open sourced github.com/41fred/claude-…

@malkoun_anthony @levelsio great point though! i'll update the repo with clearer instructions on this

@41frd @levelsio Do I have to have my repositories under it? O constantly have to give Claude code permission to read from elsewhere

✨ A dream I had finally came true: I can now chat directly with my sites to build any feature or fix any bug just via Telegram I've been playing with OpenClaw for 3 weeks now and it's great but I was always too scared to run it on any production server And I was right a bit as @marckohlbrugge was able to hack it by social engineering and acting as if it was me, and with enough tries it believed him, and was able to modify the server, change SSH keys etc. of course I had it isolated properly on its own VPS and it didn't touch anything sensitive (as it should!) Marc then reported that bug to @steipete who patched it fast But I wanted to try something more basic and simple, and I think maybe more secure: to just connect Claude Code on my server to Telegram which would be hard locked to only messages from me So I installed claude-code-telegram by @RichardAtCT on the server and run it as a system daemon and it works really well The cool thing is that I was already using Telegram for server errors like this: > Photo AI - ❌ Random credits giveaway failed (Attempt 30/30) with an exception: SQLSTATE[HY000]: General error: 5 database is locked So now I can just reply, "Ok fix this", and Claude Code on the server in production will try (and probably succeed) in fixing it In the video below I asked it to make show [🌳 Parks ] on the map by default on load, it did that, then I reloaded the page and it instantly worked One thing it still needs is sending actual messages while it's doing stuff which OpenClaw does really well, it's annoying to just wait while it says "Working..." but that's probably next

@malkoun_anthony @levelsio Claude runs inside that directory so it picks up the repo's CLAUDE.md automatically. if your CLAUDE.md says "this is a Next.js app, tests are in /tests, never touch .env" then Claude follows that when answering from Slack.

@malkoun_anthony @levelsio good question! no, they're completely separate. the daemon lives in its own folder and has a WORKSPACE_DIR env var that points to whatever repo you want Claude to work on. they don't need to be near each other.

@malkoun_anthony I like that.

@41frd I want to eg. talk to it while driving to check my messages, draft and send responses, check my calendar, email, all sorts of things.

Who's building a mobile app that lets me talk to an LLM connected to custom MCP servers?

@malkoun_anthony @iamramiawar @teknium Always Claude bro

@iamramiawar @teknium Have you tried the gemini pro models? I find it hard to pick between gemini and claude nowadays

still havent been able to try sonnet because cursor seems broken - wdyt?

@malkoun_anthony @FarzaTV @_buildspace That would be sick

So sad to hear about buildspace shutting down, and Sage with it😭 Is there a way we can keep Sage going? It was such a useful tool. I'm thinking of creating an open source version of it if that's alright with the team @FarzaTV @_buildspace Much love and respect for you all <3

@iamramiawar @malkoun_anthony 💯!

@41frd @malkoun_anthony exactly.. haven't thought about it from this perspective tho. another useful tool would be "un-SEOing" a blog post 😂

Google can't afford this much compute, but our computers can 😏 This morning I built a chrome extension that opens and reads each search result, summarizes contents, and judges relevancy to your initial question. ⚡⚡⚡ All offline, in-browser LLMs.🤯 Does this interest you?

@malkoun_anthony @iamramiawar This is cool @iamramiawar. Especially that most articles we read are just SEO optimized content dumps

@iamramiawar Would be nice to have it rerank the results based on a relevancy score

@malkoun_anthony @_buildspace @_nightsweekends @FarzaTV I can back this. Tested it out on an ecom dataset and it was great!

n&w people, hope I'm not late to the party! You can already go check out the project at dataline.app 😎 cc @_buildspace @_nightsweekends @FarzaTV

There is a deeper truth which I struggle to fully articulate: Truth is a story. Narrative shapes Truth. Truth flows from Narrative. "That's not true", your story says, "I have facts!" But facts alone are meaningless. The Narrative determines which facts are relevant and reliable, provides the context and meaning, labels the heroes and villains, and declares what is accurate information and what is harmful misinformation. And so we are ruled by stories, Narrative. All of our most divisive issues are ultimately a war of narrative. We may argue over facts, but the real conflict is narrative, and the struggle for who controls The Narrative. These narratives are unavoidable. We can't make sense of the world without them. But we can become aware of them, stepping back from the conflict to realize that both sides are each working from within their own narrative bubbles. The path to peace is not in domination, one narrative wiping out all others, but in understanding. We can develop a meta-narrative that recognizes and integrates all other narratives, not as true and false but more like species of bacteria, spreading, competing, mutating, colonizing minds, sometimes helping and sometimes harming. We must understand that those colonized by one or the other narrative are generally unaware, to them it is simply the truth. This meta-narrative brings awareness, and with it some degree of freedom. We can choose our narratives. We can change our narratives. And through this process of awareness, integration, and refinement, all narratives are transformed. This is the beginning of Alignment.