who the f*ck let my "reviewer" agent approve its own homework... i am never building an agent team without a lock again
2am. my agent team says "done ✓"
i almost hit publish
then i opened the work
→ 3 claims with no source
→ 1 section just missing
→ the reviewer agent had written "looks good" and closed the task
turns out the claude code docs say it out loud: the lead approves every teammate's plan automatically. nobody asks you
a reviewer that can't stop "done" is just a comment with an avatar
so i wrote a 13-line rule: no task closes until the reviewer writes PASS
→ no review file: blocked
→ "NOT READY, 2 problems": blocked
→ "PASS": task closes
13 lines, $0 a run, and "done" finally means done
the rule and all 4 agent files are in the article below ↓
One Grok agent just sat on the market lab and turned a $25,000 balance into $45,689 while its mood flipped from calm to euphoric every time the fills landed.
Forget staring at charts and second-guessing every tick. The agent reads order flow, tracks 64 markets, and lets 12 of itself react in real time. Balance moves, mood changes, equity curve keeps climbing.
You watch it start calm. Then pleased. Then focused. Then euphoric when the big numbers hit. Even the stung moment gets absorbed and the line still finishes higher.
How the run looked:
1. Starts at $25,000. Zero profit. Calm face.
2. First fills come in. Balance ticks to $25,424. Mood turns pleased.
3. Bigger pushes. +$1,840 flash. Balance $27,759. Euphoric.
4. Drawdown hits. -$620. Stung for a second.
5. Recovers. Focused again. Then another euphoric run.
6. Ends near $45,689. +$20,690 on the session. +82.8%.
Signals climb past 2,700. Order flow particles keep moving. Equity curve stays pointed up the whole time.
Most people still flip between a few charts and freeze on red candles. This agent just stays in the lab, changes mood with the market, and lets the balance keep growing.
Screen says the numbers are illustrative. The part that sticks is watching one face ride the waves and finish almost double while you barely touch anything.
One agent. 64 markets. Mood follows the money. The curve just kept going up.
The greatest minds are on X 💡
Incredible session with @advertisingweek’s Lance Pillersdorf in conversation with @Monique_X and @MorganStanley’s Wealth Management CMO Andrea Zaretsky on the rapid innovation happening at X.
i still can't f**king get why people keep adding more Grok Bots instead of copying the rules SpaceXAI runs them on
the Grok Bot team runs dozens of bots internally on 5 rules, and they just put them in the official docs
the useful part:
> one bot, one job, one owner
> do it by hand first, then save it as a skill
> automate only what already worked twice
> no fresh source, no run - report it, never use old data
> nothing gets sent, bought, deleted or posted without your approval
if a bot keeps making the same mistake, don't rewrite the prompt. turn the fix into a skill
that's basically the whole game
less prompting. better roles. better guardrails
all the rules are in the reply, and you can set them up as a skill in your Grok Bot
save it before you build your next Grok Bot team ⭣