@watanabe_kai_ Also, identifying when some is clearly a weirdo. Seems to be a lost skill now that online activity is the main social outlet for people these days.
@LeagueOfLegends@RiotPabro@RiotMeddler@RiotPhreak@RiotFeralPony Everybody is killing it. Content, PR, Social, and of course, the game team. I haven’t been this hyped to play League in some time.
There is only one thing that could possibly make me even more excited….
@DShyKnock To the Saudis, maybe. There is no money in esports outside of it. Both orgs are flat broke. It’s not ego anymore, they would sell if they could.
@cjzafir@rishflips Good info, but don’t do it this way. Either route through MCP or have Claude literally launch Codex in terminal. This way you can retain skills that you’ve installed directly in Codex.
Pro tip: this also works with Composer/Grok/GLM/any other terminal based LLM.
@LiuLawrence45@TheFloridaMan@5harath I was super skeptical, but I reviewed your policies, found nothing, and decided to give it a try. Easily the best dictation app I’ve ever used, nothing comes close.
@yordisprieto@BHolmesDev Do you see a distinct benefit from generating the plan with Codex vs Opus?
Seems like an unnecessary extra step. Hard agree with 3 though.
/goal is now my default mode of working with agents. but I spend a ton of time preparing all of the context and rubric criteria before kicking off the job
imo every agent message that’s not attached to a /goal should be in service of generating a spec + rubric to launch with /goal
because:
1. “verifiability” of a dispatched sub-task is the largest determinant of an agent successfully completing it
2. “complete this task to completion” is everyone’s desired UX, no one wants a half done solution, that’s what /goal tries to give you
3. thinking in terms of verification and units of work is a good mental model for thinking about sending agents to do useful things
@jxnlco Yes, sometimes, and I actively pick based on reasoning level. For example, if I’m delegating firm requirements to Codex for coding, I’ll use low reasoning. If I need it to think, it’ll be high or extra high.