trevin@trevin7h ago

Compound Engineering 3.26 is out! 🌐 New website making us look more professional 😛. https://t.co/YmqTzRGvgC 🎨 📝 ce-prototype now supports annotations We added our prototyping skill last month and have now expanded it. When web prototypes are created, you can interactively annotate with comments and send the set to the agent, who will then iterate another turn. It's a very fluid way to iterate quickly, especially for nuanced feedback on UI elements or interactions 🥖 Multi-model design bake-offs We've added a new ce-bakeoff skill that can be used to get multi-model proposals on architecture or design approaches. ce-plan will automatically invoke a "bake-off" with 3 different models to get proposals, then the orchestrator picks the best one or takes pieces. This will meaningfully improve your outcomes automatically. It works when you have multiple CLIs avail or working in harness that supports multi-family models like @cursor_ai. 🫠 Fight slop Latest frontier models like Fable 5.1 and Astra are better for writing, but still have writing slop galore. ce-noslop is our anti-AI writing slop skill, which the rest of our skills use to make communication with you clearer and the writing better. You can also use it on demand in your own workflows for any writing you're doing. Note: the skill's goal isn't to be personalized to emulate your writing style, it's just to remove AI slop bullsh*t. You could use ce-noslop as an ingredient to a larger personalized writing style skill ofc. Funny enough, I didn't use it to author this post. I tradAuthored it like a savage. 🤏 Right-sized code reviews Code reviews are better sized for the work it's evaluating. Improved it's ability to discern when changes are smaller, which doesn't warrant over-review (and the token consumption that goes along with it). We always apply your coding standards in CODING_STANDARDS.md if it's present ofc. 🚀 LFG got more flexible lfg is our skill that was used to automate the workflow taking you from plan->PR. We've made it more fully an "autopilot" by interpreting your prompt and activating the right CE skills. E.g. `/lfg fix this bug ABC-123` will auto invoke ce-debug. `/lfg let's brainstorm a new feature to improve sign up` will auto invoke ce-brainstorm. etc etc. 📦🧪 Packs (experimental) 3.26 introduces a new experimental feature called "Packs" to level up your workflows. It's such a big feature, I can't do it justice in a bulleted tweet. More on that soon. But if you're interested you can read about it here: https://t.co/BTHxbsAJnu. Warning: given it's new, it's shape might change.