another development :)
my samsung TV loads a region specific convolutional neural network (LogoCNN.bin, cnn.mapping) which tracks your usage in app.
like in youtube it's looking for videos you watch, ads you skip, progress in a video, etc.
so glad they're interested in me.
BBC: Hablaremos con nuestro experto en seguridad de BBC respecto a la vulnerabilidad de PHP.
El experto en seguridad: No sabía que PHP existía hasta ayer en la mañana.
i still prefer native apps.
> codex
> opencode
> claude
> agy
is what i use often
so, i dont even need to resume the same exact session.
just asking agent to resume with the session name like "resume the claude/codex/opencode/agy session - "name of the session" " does pretty much all of it.
the model finds the sessions, read it and continues.
this is why i say model eats harness
goated multi-agent coding tools you probably haven't tried:
- Orca ( https://t.co/bl9GGQKA3R )
- Herdr ( https://t.co/fS39C5nyNz )
- Paseo ( https://t.co/DjZcJY83JA )
- T3 Code ( https://t.co/vtoutunzNB )
you can now stop switching between claude code and codex all the time
@aaryabadhe i think no sugar is a bad advice if the goal is to lose weight.
a better way is restrict calories. so within your calorie budget you can have anything including sugar if you can track it.
if you’re diabetic then its a whole different topic altogether.
@catmanyau i think all of them won’t disappear but what i think is no harness would be better than other as models get better just because of a special tool or special or optimised session management.
@catmanyau sessions are dependent of harnesses.
so codex sessions are under ~/.codex
claude under ~/.claude
and the model knows about it. that is what i take advantage of.
you are right, no shared transcript
I tested agent-browser vs browser-use (CLI) on real-world tasks.
agent-browser: 15/15 successful runs
browser-use: 12/15 successful runs
agent-browser was also faster on every task both tools completed.
If you're giving AI agents browser access in production, this comparison might save you some time.
Full benchmarks: https://t.co/lLPHGBAFMc
for some reason executives get really excited when they hear ‘multi-agent …’
using this one agent loop seems pretty boring
but is exactly what gets the job done.
i even don’t find using opus for planning and sonnet for writing code a good option.
a better idea is simply lower the reasoning after the plan because if not opus will again verify sonnets task and so on