@denismuteguya2@Ripdizzle1@grok@BrianRoemmele if you asking this question here instead of using AI to answer your question, you should forget about AI you are to dumb, go do something else.
@sukh_saroy to be honest if you need to use 1000 repos, skills, harnesses etc, to make LLM to do what it should do out of the box, for me this means the product itself is f.. up big time, is just not working
@Bhavani_00007 if this was true this would be great, but claude is far away from this ... is lazy, is lying, makes tons of errors in unfinished code ... do good luck using claude
@Amank1412 because anthropic wants you to want it so much , so when they release it to peasants, you will beg on your knees for it, for any price, whatever they will give you, even shit on the stick, you will be so happy you finally got it.
@VadimStrizheus new sessions - use them regularly against your intuition, use caveman skill (google it) create persistent memory - check google claud tech yt - cancel claude move to codex
@melvynx open claw bots like - do lots of crazy stuff in background, the are like crazy cars running arounds town burning petrol. Just ask yourself would you do this if you had a car, just run around or would you drive to places that make sense and so you do not waste gas?
@danveloper start using new session for everything you will save tons of tokens, ai has less to read each time you type new message. research this topic
@SinitskiM@TeeDevh ask you claude code to evaluate code base and then ask claude to evaluate its evaluation and see what you get and do it again .... my experience - garbage unless you have really solid harness, lies/laziness, is like fixing claude code for anthropic for free, separate job