Monday edition. It reviews the full week Mon Sep 28 → Sun Oct 4 (7 UTC days) against Sep 21–27. Solana volume was $18.0M vs $4.66M (+286%); $12.9M of it from ordinary wallets (+381%). Saturday's deck said the post-spike days were "a new baseline ~6× the old one". The weekend answered: Sat $280K, Sun $231K excl. high-frequency, traders 1.6K, the same as any day before $BAGSPAY. $BAGSPAY itself went $2.77M (Tue) → $47K (Sun). The good news is structural: public revenue was $82.1K (+84%) and, for the first time, the on-chain take table reconciles with it to within a few percent (take rate 0.45%). Eng shipped the rest of Bags Payout over the weekend (backend payout-bot integration #1356, cashback claimables #1358, BagsPay real send #553) while 99.8 SOL of routed creator fees sits unpaid in the @BagsPay wallet and the accounts stayed silent a second week. Robinhood: 35.5 ETH ≈ $96K for the week (-81%), 23 launches; the vault is at zero. Our own brrr.trade launcher produced 79% of "launches" and 196 stock pairs that traded $15K in total.
Operating frame: Bags runs as a KPI-focused firm aimed at capturing trading volume through mobile and terminal products. The stats come prefilled (this brief is the prefill); the meeting discusses why. This week's why: volume follows payouts, and payouts stopped on Oct 1 when the BagsPay bot was handed over. The replacement merged Oct 4. The weekend was the gap.
Key stats: Solana $18.0M (+286%), 31.1K traders (+219%), 689K trades, 2,395 launches (+20%) from 268 creators (+50%). Creator claims $144K (+84%). Platform take $80.6K / 678 SOL (+105%) = public revenue $82.1K on DefiLlama (+84%). Robinhood 35.5 ETH ≈ $96K (-81%), 23 launches; free gas ended Sep 29 and the vault reads 0.0 ETH. Eng: 176 human commits (−38%), 37 PRs merged, Linear closed 10. Headline: the best full week since summer, carried by one coin, and it was over by Saturday.
Solana by day (excl. high-frequency wallets): $1.95M → $5.93M → $1.48M → $1.71M → $1.28M → $280K → $231K. Daily traders 5.8K → 16.5K → 6–7K → 5K → 2.2K → 1.6K. $BAGSPAY was $5.1M of buy-side volume (337K trades, 12.1K makers) and faded to $47K on Sunday. Launches are the wrong metric and this week proves it: 53 wallets launching ≥20 times made 1,904 of the 2,395 (79%), and a good share of those are our own brrr.trade deployers (Oct 3: 662 launches from 36 creators, 152 of them stock pairs, when the stock lane was refunded). Organic: 215 creators (+78%) launched 491 coins (+160%), the real number. Stock pairs: 196 launched, $15.3K traded, 425 buyers. Supply nobody wants yet.
Robinhood: 35.5 ETH ≈ $96K over Sep 28–Oct 4 vs 184.5 ETH for Sep 21–27 (-81%), 23 launches vs 64. By day: 7.0 → 20.7 → 2.9 → 2.1 → 0.7 → 1.0 → 1.1 ETH. Curve 22.1 / pool 13.4 ETH, 448 buys / 449 sells / 382 pool swaps. Fees (2%) ≈ 0.71 ETH ≈ $1.9K, protocol half ≈ $962. All-time launches 4,755 (+82 on the cache refresh). Vault 0.0 ETH. Post-subsidy RH is a tail until stock pairs give people a reason to be there.
Bags Payout is now complete in the core stack. Over the weekend ramyodev merged backend #1356 (payout-bot integration), #1358 (cashback vault deposit, claimables and claim-tx endpoints) and twitter-claim #553 (BagsPay real send, cashback claims, Bags payout), on top of last week's #1351/#1355 mapping + read endpoints and finnbags' #550/#552 launch-form fee sharing and /pay. #549 (token-page claim status, +10.6K) was updated this morning and is the next merge. Mobile has now been dark two weeks: 0 bagsapp commits; the only mobile PR is danbagsfm's Relay cross-chain (#690). Indexer/datapi default branches were also 0, with alaa's Postgres shadow store (#161/#152) still in review.
Take rate is reconciled. The platform-take table and DefiLlama daily revenue agree every day this week (Sep 29: $47.8K vs $48.5K; Oct 1: $9.7K vs $9.9K; Sun: $1.08K vs $1.13K). The "dollar figure, not a rate" flag carried since Sep 14 can be retired: take ≈ 0.45% of volume, half of the ~0.9% fee. Cantina: 7 closed (nikola), 3 in review (BAGS-467/479/483), 16 new Low follow-ups filed (BAGS-471…486); the three High items (BAGS-443/442/440 secrets, IAM, Flux pins) are still untouched. Ops: dividend-bot moved to k8s (#9, sadj). RH scanner: week-scan2.ts was dropped for week-scan-window.ts with explicit START/END, which refreshes the token cache first.
Three items that can't wait. (1) 99.8 SOL in the @BagsPay wallet. Fees have routed there at 100% since Sep 28; the fbmini bot was handed to Ramy on Oct 1 and nothing has paid out since. Holders on X are asking "awake soon?" and "kill us or save us?". Now that #1356/#553 are merged, the first backend-driven payout is the receipt. (2) Finn's Oct 4 call: stop our own launchers routing fees to BagsPay. Mixing house launches into the holders' pot muddies both. (3) Which BagsPay coin is "official". The dominant X thread is CA confusion (H8WT… vs J9kw… vs BAGSPAYBOT) plus an affiliate post quoting a $1.2M MC. One sentence from the main account fixes it. Close with named owners.
176 human commits on default branches (282 the week before, −38%), 37 PRs merged (31 by humans), and Linear closed 10. Five humans committed versus eight. The drop is partly real (Wed Sep 30 had 3 commits) and partly where the work lives: alaa and dan are on feature branches (#161, #541/#690) that show up as PRs, not commits.
Human commits on default branches, Sep 28 → Oct 4 UTC (deploy[bot], bags-deploy-bot, github-actions, renovate and cursoragent excluded). Prior week: ramyodev 93, finnbags 92, alaabags 30, unbound-bags 22, sadjbags 14, danbagsfm 14, ramy-zemo 12, nilsbags 5 = 282. Linear: 10 closed — 7 Cantina fixes (nikola), 3 ops/BD items (stu). Nothing closed Sat–Sun.
| Theme | What landed | Owner | Repo |
|---|---|---|---|
| 💸 Bags Payout / BagsPay — complete | Weekend merges: payout-bot integration (#1356), cashback vault deposit + claimables + claim-tx endpoints (#1358), BagsPay real send + cashback claims + Bags payout on web (#553). Earlier: token mapping and read endpoints (#1351/#1355), public-api exposure (#75/#76), fee sharing on the launch form (#550), the /pay send form (#552), receipt-style claim history (#548). | ramyodev · finnbags | bags.backend · twitter-claim · public-api-v2 |
| 🚀 Launch form + profiles | Launch-form redesign with a fee-sharing app rail, Bags v3 Robinhood launches, BagsPay palette and home refresh (#538/#542/#543, ~31K lines), release #544 + bugbot follow-ups (#545), profile OG wordmark (#546/#547). | finnbags | bags-twitter-claim |
| 🛡 Cantina security fixes | Proxied user from the forwarded API key (#1307), auth code claimed once before MFA (#1310), private IPs blocked in metadata fetch (#1312), API keys refused for banned users (public-api #69). 7 Linear issues closed; 3 more in review. | unbound-bags · nikola | bags.backend · public-api-v2 |
| 📐 Data accuracy | Swap, P&L and points record accuracy (#1347); earnings, market cap and claim labels (#1349); pre-launch tokens hidden from claimable positions v2 (#1342). | unbound-bags · ramyodev | bags.backend |
| 🧹 Launch pipeline + ops | DAMM v2 direct launches removed (#1345), public-API launches tagged (#1354), contract-failure-rate alert (#1350), Mongo indexes on startup (#1352), connect-wallet add endpoint removed (#1344), dividend-bot on k8s (#9). | ramyodev · nilsbags · sadjbags | bags.backend · dividend-bot |
Read: the product that drove the volume is now fully in the core stack, including the weekend. The risk is unchanged: finnbags + ramyodev = 160 of 176 commits (91%), and Ramy now owns payouts, cashback claims and the BagsPay cutover at once.
Same three clusters as last week: (1) Relay cross-chain on web, backend and mobile (+36K across three PRs); (2) alaa's Postgres current-state shadow store (+26K across indexer and datapi); (3) unbound-bags' hardening follow-ups. New this week: #549 got its update this morning and the stock quote catalog refresh (#160) moved, which the stock-pairs Linear items depend on.
| PR | Scope | Owner | Status | |
|---|---|---|---|---|
| HOT | twitter-claim #549 — token-page claim status + phone layout | +10,582, updated Oct 5. Shows whether a coin's fees are claimed, plus not-found pages and launch-form polish. This is the on-site answer to "is this abandoned?". Merge it this week. | finnbags | open |
| HOT | indexer #161 · datapi #152 — Postgres current-state shadow store | +23,863 / +2,522. Shadow sink plus client and repositories. Infra that eventually makes this deck's stats cheaper and more trustworthy. | alaabags | in review |
| HOT | twitter-claim #541 · backend #1340 · bagsapp #690 — Relay cross-chain | +23,385 / +8,937 / +4,083. Trade any chain from Bags, USDC settlement. The only mobile work in flight. Still competes for review with the payout cutover. | danbagsfm | in review |
| HOT | backend #1341 · #1346 · #1348 — hardening + consistency | Low-severity security hardening (+1,113), stale caches (+416), swap-quote / claim / referral edge cases (+358). Map to Linear BAGS-467/479/483 (in review). | unbound-bags | in review |
| DATA | datapi #151 top revenue tokens · indexer #160 stock quote catalog | A top-revenue-tokens view (+1,769) and the refreshed stock quote catalog (+386, Oct 5), a prerequisite for BAGS-434/436. | nilsbags · alaabags | open |
| STALE | 5 mobile PRs untouched for 3+ months — oldest bagsapp #602 (Jul 2025) | #602 auth fixes, #609 feed charts, #618 Nativewind (+6.9K), #648 dead-code cleanup, #649 Cashback/Hyperliquid (+25.6K). Mobile's open queue is all zombies. | ronickg · danbagsfm · finnbags | 4–14 mo |
backend (19): #1356 payout bot · #1358 cashback claimables · #1355/#1351 mapping + endpoints · #1354 tag public-API launches · #1350 failure alert · #1345 remove DAMM v2 · #1342/#1344/#1352 · Cantina #1307/#1310/#1312 · data #1347/#1349 · 4 releases. twitter-claim (11): #553 real send · #552 /pay · #550 fee sharing · #548 receipts · #538/#542/#543/#544 launch form · #545–547. public-api-v2 (6): #69 · #70 · #71/#73 · 2 releases. dividend-bot (1): #9 k8s.
Three +20K PRs are open at once (Relay web, Relay backend + mobile, shadow sink) and all three are over a week old with no merge plan. Last week's sequence still stands: payout cutover (done) → #549 → shadow store → Relay. Also: 35 open PRs is flat on last week, which means the merge rate equals the open rate. Anything over +5K needs a merge date within two weeks, or it gets split.
Default-branch commits only, 14 repos. Indexer and datapi work is in the open shadow-store PRs (#161/#152). bagsapp: 0 commits for the second week; its only live PR is Relay #690.
The 10 closes are the same 10 as Saturday's deck; nothing closed over the weekend while 4 PRs merged. 24 issues were created (16 are Low-priority Cantina follow-ups BAGS-471…486, four are stu's travel todos). 39 of 49 "started" issues haven't moved in 14+ days. BAGS backlog is 117 (14 High, 0 Urgent). The payout cutover, the launch-form redesign and the cashback claims work still have no Linear issue. The stock-pairs cluster (BAGS-434/436/426/412, all High) has been untouched since Sep 14, the same week we launched 196 stock pairs that traded $15K. Linear describes a different company than GitHub does.
Solana did $18.0M vs $4.66M (full week). Excluding high-frequency wallets (≥200 trades in the window), $12.9M vs $2.67M (+381%). The spike was $BAGSPAY on Tue Sep 29; Wed–Fri held $1.3–1.7M; then Sat $280K and Sun $231K, inside the pre-spike range ($190–280K). Saturday's "6× floor" call was five days old and wrong. Robinhood: 35.5 ETH ≈ $96K (-81%), post-subsidy.
Ordinary-wallet volume by day: $1.95M · $5.93M · $1.48M · $1.71M · $1.28M · $280K · $231K. Daily traders: 5.8K · 16.5K · 6.0K · 6.8K · 5.0K · 2.2K · 1.6K. $BAGSPAY buy-side: $462K → $2.77M → $1.15M → $380K → $238K → $88K → $47K. The weekday numbers were real and they were one trade: buy the coin whose fees pay out. Payouts stopped Oct 1; the volume followed three days later. The number to watch this week is Mon–Tue ordinary-wallet volume: above $1M says a habit, below $300K says it was the coin.
Platform take vs DefiLlama revenue, by day: $2.7K/$2.9K · $47.8K/$48.5K · $11.4K/$11.4K · $9.7K/$9.9K · $6.7K/$6.9K · $1.3K/$1.3K · $1.1K/$1.1K. Week: $80.6K vs $82.1K. The accrual table has been consistent with the public basis for seven straight days, so the "dollar figure, not a rate" caveat from Sep 14/21/28 is closed. Take rate this week: 0.45% of volume (prior week 0.85%, mix-driven: the DAMM v2 leg was $14.0M of the $18.0M). Quote it as a rate again.
Net read: we still own the idea in public and we are losing the holders in private. Last week's deck said the answer was a receipt. This week the receipt has a number on it: 99.8 SOL sits in the @BagsPay wallet, unpaid since Oct 1, while the code to pay it merged Oct 4. Pay it, show the tx, say which coin the account stands behind, and the three bear threads above close in one post. Keep waiting and the Chinese CT thread telling our origin story becomes the obituary.
#1356 (payout-bot integration) and #553 (real send) merged Oct 4. The @BagsPay wallet holds ~99.8 SOL (≈$12K) of creator fees routed at 100% since Sep 28, untouched since the fbmini bot was handed to Ramy on Oct 1. Owner: ramyodev. Steps: dry-run the payout bot against the wallet, pay out, post the tx. Also action Finn's Oct 4 call: stop brrr.trade launchers routing fees to BagsPay (flip HOUSE_FEE_RECIPIENT) so house launches stop mixing into the holders' pot. File it in Linear; it still doesn't exist there.
Fourteen days without a post from any official account. The post: $144K paid to creators last week, $82K protocol revenue, the payout tx from objective 1, and where to check claims (/pay, claim history, #549 once merged). Add one sentence on what @BagsPay is and which coin, if any, the account stands behind, and that Bags accounts don't quote market caps. No competitor names, no squatter CAs. Approval still required; the Sep 29 draft thread is the base.
This week 53 wallets made 79% of launches, and the Oct 3 record (662) was our own stock lane. Public stats, the home page counters and this deck should report organic creators (215, +78%) and ordinary-wallet daily volume, and the brrr.trade deployers should be tagged (#1354 already tags public-API launches; extend the tag to known house wallets). Set the floor KPI: ≥$1M/day ordinary-wallet volume and ≥5K daily traders. Sat/Sun were $280K/$231K and 2.2K/1.6K; Monday tells us if the weekdays were a habit or a coin.
196 stock pairs launched this week traded $15.3K across 425 buyers, while the stock-pairs product work (BAGS-434 DBC curves, BAGS-436 QA, BAGS-426 dashboard, BAGS-412 terminal filter) is untouched since Sep 14. We are manufacturing supply for a product that isn't built. Pause the brrr.trade stock lane (keep the kicker idle), ship BAGS-434/436 with an owner, then turn it back on. Hold Relay (#541/#1340/#690, +36K) until #549 and the shadow store are in; three +20K PRs in review at once is how the bus factor bites.
#549 (token-page claim status, +10.6K, updated this morning) is the on-site answer to "abandoned?". Merge it by Wednesday. Mobile has had 0 commits for two weeks and five zombie PRs back to Jul 2025. Decide: either dan's Relay #690 is the mobile roadmap and gets reviewed, or the mobile queue gets closed and mobile is declared maintenance-only for Q4. Not deciding is the expensive option.
Read: Linear tracked the audit (10 closes, 16 new) and nothing else that mattered this week. Same gap as the last three decks.
Not auto-created. Say the word and I'll file these with teams and priorities.
Last week's calendar had Monday as "the receipt". It didn't go out, and the weekend volume shows what silence costs. Same rule, higher stakes: every post links to something verifiable (a payout tx, /pay history, the claim status on the token page). No CAs for coins Bags didn't launch, no market caps, no competitor callouts. Every post here is a plan; Finn approves each one.
Cadence rule: MON (clarity) and TUE (the tx) are non-negotiable. Then a weekly receipt every Monday, so "are they still paying?" always has a dated, public answer.
Goal: $100M cumulative protocol revenue. Banked all-time: $32.18M (DefiLlama, Solana basis; Robinhood revenue is additive and not included). That's 32.2%. This week added $82.1K (+84% WoW), the best full week since the summer, and the on-chain take now matches it, so the number is trustworthy. It is also ~1/28th of the ~$2.3M a week the Q4 target needs with 12½ weeks left. The slope changes when weekday volume survives a weekend, and this one didn't.
| Milestone | Cumulative revenue | Increment | Status |
|---|---|---|---|
| Today (Oct 5) MEASURED | $32.18M | +$82.1K last 7d (Solana public) | 32.2% of goal |
| True incl. RH EST | $32.18M + RH (not re-summed this run) | +≈$962 RH protocol half (7d) | DefiLlama can't see RH |
| End of Q3'26 (Sep 30) TARGET | $36M | — | Missed: closed at ~$32.1M |
| End of Q4'26 TARGET | $61M | +$28.8M | needs ~$2.3M/wk · doing $82K/wk |
| $100M GOAL | $100M | +$67.8M | Needs a floor, not a spike |
Basis: DefiLlama protocol revenue, Solana (all-time $32.18M, fees $64.38M; 7d $82.1K revenue / $164.2K fees; 30d $199.0K). The honest read: $BAGSPAY week moved the bar by 0.08 points. The Sep 28 deck said the floor was the result; the weekend says there is no floor yet, only a trade that works while payouts run. The milestone to measure against is unchanged: $1M/day of ordinary-wallet volume that survives a weekend, with payouts as the mechanism. At this week's reconciled 0.45% take, that floor is ≈$32K/week of revenue; the Q4 target needs ~70× that, which is the gap between "Bags has a product" and "Bags has a market".
bags_trade_feed, UTC days, excluding wallets with ≥200 trades in the two-week window. Including them: Tue 29 = $8.72M and the week = $18.0M. Grey = the week before; red = the weekend, back inside the pre-spike range.
On-chain window scan, chain 4663, exact UTC days Sep 28 → Oct 4 (cache refreshed to block 80.99M, +82 tokens, 4,755 all-time). Week = 35.5 ETH ≈ $96K, 23 launches, vs 184.5 ETH / 64 launches the week before (Sep 28 deck scan) (-81%). Free gas ended Sep 29. Faded = the week before.