Martin Kessler
1K posts
Martin Kessler
@kesslerIO
Co-Founder/COO of 3D body tracker @shapescale. 💪 Host of 20 Minute Fitness podcast.🎙️Sports nutritionist, fitness tech geek, dad and husband. 🏃👊🔥
San Francisco Bay Area 가입일 Şubat 2009
701 팔로잉911 팔로워
Yes, the lab leak theory turned out to almost certainly false.
English
Ok, it's available:
openclaw plugins update openclaw-codex-app-server@beta
It should install v0.4.0-beta.2
When you run `/cas_resume` it pins a version of the `/cas_status` message which gives you a button to toggle from Default Access to Full Access. Additionally you can run `/cas_resume --yolo` to go directly into Full Access mode. All tested on my end and working. It's rather complicated because it appears to require having two instances of Codex App Server, one for each permission mode and moving the threads between them when permissions change.
English
@openclaw Codex App Server - Your bridge to using Codex in OpenClaw
youtu.be/GKkipfNEJJQ?si…
YouTube
English
@kesslerIO @openclaw Was just working on that. Had not actually implemented that one yet 😂 I have it spec'd out. Should be a couple hours tops.
English
@yashhsm Came here to disagree, but nah you're right! It takes a ton of work to get to a reliable workflow. In the end, you end up needing to build a lot of deterministic guardrails to make it work reliably. Looking at the past 12 months, I can see agents soon getting better at this tho.
English
I won't stop until all PDFs in the world are converted to MDs
Oliver Prompts@oliviscusAI
someone just open-sourced a tool that converts pdfs to markdown at 100 pages per second. 100% free. runs entirely on cpu. no expensive gpus needed.
English
@stratechery We're probably fine for now. But once your barber starts talking about his agentic workflows, orchestrator, and skills he's created, that's when we're truly in a bubble. 😅
English
Agents Over Bubbles
Agents are fundamentally changing the shape of demand for compute, both in terms of how they work and in terms of who will use them. They're so compelling that I no longer believe we're in a bubble.
stratechery.com/2026/agents-ov…
English
AI is supposed to save me time, but now I find myself building stuff all evening and weekend and it's actually increasing my time in front of the computer
WTF
English
While my attempts at AI automation often feel like they take up more time than it actually saves me, it's also an incredibly addictive experience.
English
@garrytan So does this replace the standard plan exit review or is this for broader/complexer issues?
English
I took my normal "exit plan mode" skill and expanded it into a "mega exit plan mode" skill that I find can be quite useful when you are trying to expand scope and/or just want Claude Code to delight you with new ideas.
It's sort of like the expansive CEO to the normal skill's realistic eng manager persona.
They work well back to back: mega skill to be expansive, normal skill to think through the final details.
gist.github.com/garrytan/120bd…
English
@thisiskp_ Switched to Codex 5.3 for everything but I miss Opus and Sonnet 🥲. Kimi has shitty privacy terms and still makes too many mistakes but I use it for simple sub-agent research tasks. Minimax seems to be overrated in real use.
English
@steipete Super annoying. You know ClawHub could be the app store for agents, but discovery is still a PITA.
English
Finally have quite good virus/malware protection on ClawHub, now these folks moved to scam in comments.
Can't have nice things. 🙃
English
@anothercohen The Mac app still sucks. Great features but feels buggy and sluggish.
English
Does anyone run into issues using connectors in Claude and it constantly timing out?
Claude Code works fine but Claude Desktop not working well for me
English
@nateliason Minimax and Kimi are both kinda okay but lack personality and when it comes to coding or true agentic stuff they still fall short. I mean if you don't mind checking on them or needing a few attempts they can absolutely get the job done. But also is it worth the privacy trade off?
English
I haven't used a non-Opus model on OpenClaw in the last week or so.
Which cost-effective one is everyone liking now?
Minimax 2.5?
Kimi 2.5?
English
Felt inspired by this list and had my agent create this site: gigafucked-landing.vercel.app
Tenobrus@tenobrus
gigafucked: - grammarly - calendly - miro - retool - webflow - langchain - writer - harvey - glean - expedia - monday fucked: - accenture - intuit - notion - jasper - canva - alphasense - postman - airtable - talkdesk - sierra - zapier - replit - solace probably fucked: - cursor - pilot - clay - mercor naively seems fucked but so competent / plugged in they seem to be figuring it out on the fly anyway: - linear
English
@tenobrus Love this list. Could be much longer yet. AI is truly disruptive and yeah it's so fast changing that early gen AI apps are already getting disrupted.
English
gigafucked:
- grammarly
- calendly
- miro
- retool
- webflow
- langchain
- writer
- harvey
- glean
- expedia
- monday
fucked:
- accenture
- intuit
- notion
- jasper
- canva
- alphasense
- postman
- airtable
- talkdesk
- sierra
- zapier
- replit
- solace
probably fucked:
- cursor
- pilot
- clay
- mercor
naively seems fucked but so competent / plugged in they seem to be figuring it out on the fly anyway:
- linear
English
@LukasHozda Since when does OpenClaw need a Mac mini as a minimum requirement? Nonetheless, I'd be intrigued about what $5 hardware use cases there are?
English
Imagine that your implementation is so bad that it inspires a whole broad community to make reimplementations orders of magnitude faster
English