i need codex, but: - codex and chatgpt are one combined app (for iOS and Mac) - i get my own VM tied to my subscription, so I don't have to SSH into my mac - the VM has the memory and self-improvement of OpenClaw. it can do/become anything I ask it to - i can bring all this context and plug it in and out of any service on the web ("Sign in with OpenAI") i'd pay a lot of money for this

What started as building a personal taste.md skill for myself, turned into building a pipeline to create any taste as a skill. The most important piece is references. This is where you should spend time. If the references suck, so does the skill. I find that references cropped tightly on details in high resolution work the best. Each image gets analyzed by both Opus 4.7 and GPT 5.5. The analysis is based on why the reference is successful as a piece of design - not what it does functionally. Using two models helps rule out biases and gaps from each. The models focus on layout, spacing, typography, rhythm, composition, hierarchy, etc. At the end, each image has: reference-01/ - opus-4-7-analysis.md - gpt-5-5-analysis.md Then we fuse them together using GPT 5.5 - but the md files are anonymized so 5.5 doesn't prefer itself. reference-01/ - fused-analysis.md reference-02/ - fused-analysis.md etc. After fusion, we have one synthesized analysis per reference. Now the goal is to combine all of those into a single rule set. This is where chunking matters. If you ask one model to combine 100 image analyses at once, the result becomes too broad. It summarizes instead of preserving the granular design rules we want. Instead we chunk the fused analyses into smaller groups. Each group gets merged into a chunk-level synthesis, usually from around 6 to 8 image notes at a time. Then one final model pass fuses those chunks into a single md rule set. Finally, using the rule set, we write a skill of concrete instructions. It enforces constraints, uses imperative wording, and avoids vague taste words.

This is fully realtime — not sped up. I’ve been experimenting with Realtime-2 and trying to move beyond “turn-for-turn” conversations. No toggling Whisper. Just endlessly talking into the terminal. Paired with Codex Spark, the immediate feedback starts to feel pretty magical. By EOY, I imagine we’ll be doing this with top-tier models.