Onur Solmaz
@onusoz
Followers
4K
Following
6K
Media
207
Statuses
1K
Maintainer @openclaw, Founding Engineer @textcortex. Trying to define Agentic Engineering
Joined December 2022
The models, they just wanna work. They want to build your product, fix your bugs, serve your users. You feed them the right context, give them good tools. You don’t assume what they cannot do without trying, and you don’t prematurely constrain them into deterministic workflows.
7
3
24
wow
Claude code source code has been leaked via a map file in their npm registry! Code: https://t.co/jBiMoOzt8G
1
0
8
next up: claude agents sdk supports openai responses api 💀
We’ve seen Claude Code users bring in Codex for code review and use GPT-5.4 for more complex tasks, so we thought: why not make that easier? Today we’re open sourcing a plugin for it! You can call Codex from Claude Code with your ChatGPT subscription. We love an open ecosystem!
0
0
14
yes
learning from openclaw: 95%+ of agentic coding will be done via voice, from our phones in the future 😂
0
0
7
OpenClaw saw over 1000 PRs opened per day in early March Stuff is crazy, I wonder whether we will see 10k per day one day...
8
1
30
Here is the spec and implementation for this flow. The mermaid diagram includes all the steps I mentioned in the post above, including a shameless AI review ralph loop, and other loops to make CI pass, resolve conflicts and so on I would recommend reading the README and
0
2
57
acpx v0.4 ships Agentic Workflows, or as I like to call them "Agentic Graphs" It let's you create node-based workflows on top of ACP (Agent Client Protocol), to drive any coding agent (Codex, Claude Code, pi) through deterministic steps This let's you automate routine,
73
100
1K
OpenAI early 2020s: "This model is too dangerous to release publicly, the world is not ready for it 😱😱😱" OpenAI and Anthropic in 2026: "Anybody can code now for just $200 per month. Oh btw our models are also leet uber hackers which can find zeroday exploits in any software,
0
1
8
Wow even I as a frontend noob understand the significance of this Some distant memory from 15 years ago needing to measure the width/height of some text and finding out it’s not possible to do reliably in web More beautiful typography for the web!
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
0
1
5
There is an economic theory waiting to be uncovered here Token Leverage (TL) = Token spend / Human labor spend The higher Token Leverage a company has, the more automated and productive they are If you have TL=1, you are spending as much money on AI as your human employees
By the end of 2026, I predict token spend will be greater than engineering salaries at early stage startups.
2
1
8
Just shipped @openclaw Codex App Server plugin. Control Codex from Telegram and Discord - resume threads, switch models, approve commands, plan, review, or go full yolo mode, all from your phone. One install. Zero config. Full control. 👉 https://t.co/XVuZDeLgkP
47
55
703
There is a desperate upcoming need for version controlling non-dev knowledge work. Git for non-devs. Otherwise non-devs won't be able to use agents to their full extent Non-dev knowledge work is notoriously bad at being version controlled. You cannot UNDO edits to all MS word,
5
1
27
OpenClaw 2026.3.24 🦞 🔌 Improved OpenAI API: talk to sub-agents with @openwebui 🎛️ Skill & tool management Control UI 🎨 Slack interactive reply buttons 💅 Native Microsoft Teams 🧵 Smart Discord auto-thread naming Any client. Any model. One runtime.
github.com
Breaking Changes Gateway/OpenAI compatibility: add /v1/models and /v1/embeddings, and forward explicit model overrides through /v1/chat/completions and /v1/responses for broader client and RAG com...
264
256
2K
Another one, call me stupid: “How would Google have done it?”
This is unscientific, but there are certain keywords and phrases I use a lot while using certain models like openai's. I use them a lot because they get me what I want immediately: - plainer lang - cutover - elegant and production ready - holy grail What are yours?
2
0
10
The MCP versus CLI argument should be reframed as Computer vs No-computer argument I personally get the dunk on MCP. It didn't work last year, with earlier models. Then we saw CLIs perform much better with the same models. And giving access to bash was much simpler! Models'
20
1
36
I just registered for an .agent domain and joined the .agent community! @dutifulbob will have bob.agent if it passes :) https://t.co/lhK5MQS1sk
@agentcommunity_
agentcommunity.org
Join the movement applying for a native .agent TLD for AI agents. Pre-register your name free and get priority access if the TLD is approved.
3
0
15
Sep 2021 @lexfridman podcast with Don Knuth, they also talk about OpenAI Codex (code completion model) around 33 minute mark This aged very well https://t.co/O1eTXlHTNC
0
0
2
Damn I’m gonna have to switch to teams if it goes like that
OpenClaw now has full Teams AI UX: streaming responses, AI labels, feedback with reflective learning, welcome cards, and image understanding. Built on the official Teams SDK 🦞 FYI @steipete, @BradGroux
1
0
11
Codex's long horizon task and instruction following has been the most life-changing AI feature recently It is unlocking the next level of automation for me. I can convert my own heuristics into prompts and multiply my throughput 100x Currently spending some thought on how to
3
0
16