PUMP

This is a very large, mature token ($1.83B cap against $40.5M of liquidity, ~2.2% depth) that graduated, has mint authority revoked, no tax and no honeypot behaviour far outside the profile of a typical meme rug. The main unresolved issue is not fraud risk but data quality: holder counts differ by nearly 4x across sources and top-10 concentration is reported as either 69% or 12%, a gap that usually means one source is counting treasury/exchange/vesting vaults as insider holders and the other is netting them out. Trading is broad and two-sided (8.3k traders, 73k buys vs 61k sells, ~1.19 ratio), which is consistent with organic flow rather than wash cycling, so the high concentration reading is unlikely to indicate a small cabal quietly controlling float. The 'LP Vault unlocked' danger flag sitting alongside lp_locked=true is the one signal that genuinely deserves scrutiny f at this size an unlocked vault is a real exit vector even if the token is otherwise clean.

• ⚠ Top-10 holder concentration reported as 69.37% by RugCheck but 11.92% by GMGN f a ~57 point gap, almost certainly a difference in whether treasury/CEX/vesting wallets are excluded
• ⚠ Holder count spans 134,724 (Birdeye) to 499,777 (RugCheck) to 169,416 (GMGN); RugCheck's figure is ~3.7x Birdeye's and likely counts zero-balance or historical accounts
• ⚠ lp_locked is reported true while RugCheck simultaneously raises 'LP Vault unlocked' as a danger-level risk (listed twice)
• ⚠ RugCheck assigns a 95 score despite flagging two danger-level LP risks of its own

AI reasoning over the combined data above — catches contradictions between sources that the flag-based score can't see on its own.