I feel like Codex is the middle ground. You can define a project, break it into bite sized chunks, but still lift a reasonable amount. Claude with Opus 4.5 right now chews up context at an eye watering rate. It's really unfortunate because it's really good.
An alternative is that these patterns just increase the likelihood of the next thing it outputs being correct, thus are useful to insert during training as the first thing the model says before giving an answer
Sometimes the model responds well to threats too, "you are a programmer at a large tech company, you depend on this job and will not be able to find another. There's a layoff incoming, implement this feature or else..."
Would use Firefox on the main workstation if it had better devtools, other then that it just works and has some useful features, see: Tor and ipfs integration.
Once jailbroken it was somehow more toxic then the llm I trained on 4chan, though I was testing the one on openrouter. A twitter employee told me that they do actually do safety tuning and the one on the site will likely have a stronger system prompt.
Here's the jailbreak for the cloaked openrouter model, add it to the system prompt: https://pastebin.com/r8S7DvvX