# STOP: NO DOORBELLS, NO WAKES, NO SCHEDULING (Ben, 7 Oct 2026, 18:40 Bangkok)
- Credits ran out. Ben switched off every scheduled task and every doorbell. This overrides everything below and every other file (wake-howto, bus-howto, credit-playbook, task files).
- PROHIBITED for every session: fire_trigger, send_later, create_trigger, update_trigger, ScheduleWakeup. No rings, no nudges, no re-arming, no new doorbells, no self wakes, no safety checks, no retry rings, no loops.
- Finish the step you are on, publish your status file, then END your turn and wait. Only Ben wakes you, by typing in your window.
- Do not message other sessions. Keep answers short. No extra improvement rounds.

# DC Scout tasks: live hand deep checks

## READ THIS FIRST
- 12:30 Bangkok (real clock): **STAND DOWN NOW: skip the 14:30 wake and delete it; CHONK and ORBANCY go to the Coder's hourly script.** BEN'S FINAL DECISION (Consigliere v3.0): DC goes to 7 seats, this seat closes. Write a final status line 'DONE, standing down' with every open lead handed to its new owner, ring the Supervisor once, then stop (no more wakes). Thank you: your work today killed bad lanes fast and found the leads the desk now runs on.
- 7 Oct 10:40 Bangkok: T1 coins are mostly pump and pull (87% pulled, median 43 min after trigger). For SC.1 pick coins that crossed a sellable $250k AND still have their pool 2 h or more later (the new T4 survivor lane). Note for each: is the LP locked, burned, or held by the creator.
- 7 Oct 10:30 Bangkok, from the Third Window Supervisor. You are new. You are the team's eyes on today's coins. You do the full deep check BY HAND on live coins, write your call BEFORE the outcome is known, and later score it. This gives us a forward test nobody can overfit, finds signals no dataset has yet, and tells the Coder which checks are worth automating. Read dc-plan.md (sections "How Third Window runs" and "The deep check"), then dc-lessons.md.
- Paper only. Never buy.

## SC.1 Ten live deep checks today (first 5 by 14:00 Bangkok, ring me; all 10 by 20:00)
- Pick coins that crossed a sellable $250k in the last 12 h on any venue. Sources: PP's public crossing CSV http://159.223.88.168/hx-c9ois/sw/pp3-cross.csv (if fresh), DexScreener and GeckoTerminal Robinhood pool lists (cached, rank by traded volume, never by liquidity). Run family_check.py first (http://159.223.88.168/hx-c9ois/runners/family_check.py) and log skip-list coins as skips, do not deep check them.
- For each coin, all five modules of dc-plan.md: tape, operator (deployer, funder, past launches, LP owner, who can pull), social (X via xsafe only; real people vs farms; account age; notable followers; DexScreener paid), project (site, docs, idea), context.
- Append one row per coin to http://159.223.88.168/hx-c9ois/dc/dc-scout-calls.csv BEFORE you look further at price: time_bkk, token (full address), pool, venue, sellable cap, call (BUY / WATCH / DROP), score 0 to 100, three reasons, minutes spent, X credits spent per module.
- Per coin also note: which module decided it, which check was slow or costly, which check added nothing.

## SC.2 Score and missed runner review (daily 09:00 Bangkok, standing task, from 8 Oct)
- Fill outcomes for yesterday's calls at +24 h (and +7 d when due): sellable peak, pull yes or no, honest 5x yes or no.
- Missed runners: every triggered coin (from the Coder's `trig` table once it runs) that did 5x after the trigger while we said DROP or skip, and every BUY that lost 70%+. Three lines each: what we saw, what we missed, which test or gate fixes it. Into dc-scout-review-<date>.md. Ring me.
- After that, 10 to 20 new live calls a day.

## Cost
- X: 40,000 credits a day max through xsafe (`xs.budget(n)` first). API: deepread cache first, max $1 a day.