3 ms·

Prompt engineering was deprecated around the time Opus-/Codex-class models dropped imo. Replaced by skills up until about opus 4.5/6 dropped. Pinned model to Op

by matltc 2mo ago

Prompt engineering was deprecated around the time Opus-/Codex-class models dropped imo. Replaced by skills up until about opus 4.5/6 dropped. Pinned model to Opus 4.6 slug in ENV.ANTHROPIC_MODEL, disabled the adaptive thinking thing, and saw my skills start causing degradation Only ones I use today are for very specific quirks (eg wiredtiger/mongodb 8+ incompatible with ext4/Linux 6.19+ specifically causing segfaults. Have a 20 line mongo skill that says as much. Pinned docker container to mongo 7, can prob delete it now) I spent a few days reading up on the docs for these things, hook lifecycles, tried writing a few, but they never work as documented, or the documentation changes so frequently that whatever you built is deprecated by the time you get it humming. Now if I have some non-trivial unit of work, I basically iterate on spec in plan mode then put it on auto and let it rip. Way better results with Fable. jury out on Opus 5, but no regression like 4.7/8 Usually it's just echo "do this lil thing then pr closing issue 123" --model sonnet --effort low. Works well enough, sonnet 5 low is a workhorse and quite resourceful in a good way when things go sideways; doesn't cheat its way out IME

tedbradley 2mo ago
Interesting. So you're using Sonnet 5 on low effort as the workhorse for most parts of your plan, unless perhaps there is a part of the plan you deem too complex for it? Then, maybe throw Opus 5 or even Fable 5 at it? Have you considered using Luna on max effort for implementation? There was that recent news that they tuned its code and balancing and maybe some other stuff, dropping costs and allowing Luna to run for 20% of the API cost it had just two weeks ago. Now here's the rub: Have people with a subscription confirmed Luna max drains their usage way slower than before? With those rolling windows and the opaque "pricing" associated with them, an 80% cut to GPT-5.6 Luna might now translate into an 80% cut to using GPT-5.6 Luna with a sub. Anway, with that news, I was curious if you've tried little Luna for implementation. On https://artificialanalysis.ai/ https://artificialanalysis.ai/, for its level of "intelligence," it is cheaper than even DeepSeek. I think they want people to switch over for that alluring price cut while also giving far more usage than Anthropic. Once people stay on their plan, they make most of their profits from those same users pulling out Sol.

by matltc 2mo ago

tedbradley 2mo ago

Interesting. So you're using Sonnet 5 on low effort as the workhorse for most parts of your plan, unless perhaps there is a part of the plan you deem too complex for it? Then, maybe throw Opus 5 or even Fable 5 at it? Have you considered using Luna on max effort for implementation? There was that recent news that they tuned its code and balancing and maybe some other stuff, dropping costs and allowing Luna to run for 20% of the API cost it had just two weeks ago. Now here's the rub: Have people with a subscription confirmed Luna max drains their usage way slower than before? With those rolling windows and the opaque "pricing" associated with them, an 80% cut to GPT-5.6 Luna might now translate into an 80% cut to using GPT-5.6 Luna with a sub. Anway, with that news, I was curious if you've tried little Luna for implementation. On https://artificialanalysis.ai/ https://artificialanalysis.ai/, for its level of "intelligence," it is cheaper than even DeepSeek. I think they want people to switch over for that alluring price cut while also giving far more usage than Anthropic. Once people stay on their plan, they make most of their profits from those same users pulling out Sol.