$BTC.D Bitcoin dominance hitting resistance at 20 month SMA. If it holds, do we finally get our rising tide lifts all boats fabled alt season?
Crypto performed poorly in Q4 of the last two midterm years, 2018 and 2022. In both cases, Bitcoin hadn’t yet bottomed when the quarter began.
If Bitcoin bottomed in June 2026, the cycle is left-translated. October–December would then correspond to months 4–6 after the bottom.
Mapping the same stage of previous recoveries onto Q4 2026 gives averages of +67% for $BTC and +61% for $ETH.
Be more bullish.
Marginal buyers for majors coming next 6 months
This, ramble and free associate all the thoughts. Codex incredibly good at figuring out what my intentions are.
Bro it’s June 2026. Stop hand editing your prompts. Hold down the dictation button and ramble for 10 minutes. Give the model every fragment, caveat, example, and vibe in your head. It is literally a large language model. If it’s superhuman at anything, it’s reconstructing latent intent from language.
I built /page-speed-optimizer skill today.
You run it inside a /goal. It starts by measuring the important user flows, then works through what is actually slowing them down: page loads, navigation, interactions, layout shifts, heavy assets, third-party scripts, caching, etc.
It keeps going until the meaningful issues are fixed, ruled out with evidence, or left as explicit human decisions. Every change gets before/after results, and the state + ledger track what happened.
github.com/nodatall/primedir…
Idea for Codex feature, when you submit a message in a thread, have option to send to a subagent. Like steer to subagent. @thsottiaux
After codex app updates, my /goal threads lose full access permissions and I have to re-create them. Very annoying. Also, if I ask an orchestrator thread to create worker /goal threads, those threads don't get full access even if I specify. Please fix. @thsottiaux
If you instruct a codex thread to create another thread and run a /goal, the permissions in the spawned /goal thread don't have full access even if instructed. Any way to get around this?
Managing agent loops is all the rage. I'm building an app that needs extensive data from all colleges in the US and ingests from a standardized source, CDS.
1. One agent goal runs the CDS ingestion
2. One agent goal improves CDS speed toward 4x
3. A supervisor goal watches both, restarts them if they stall, and edits their prompts if the instructions are what caused the stall.
Lastly, an automation wakes the supervisor every ~20 min to check on the worker agents.
Basically: goals do the work, supervisor keeps them alive, automation keeps the supervisor awake.
I built a writing process skill today called Clarifier. It helps you take a rough draft through a loop of feedback and rewriting. The agent names the top issue, asks clarifying questions, and offers short rewrite examples. Then you write the next draft yourself and repeat the loop. The skill is designed to produce stronger, more polished work while improving my writing skills.
github.com/nodatall/primedir…













