Blockchain Bets ERC-1155 stake/transform inflates redeemable BCB → UniV2 dump

Source & credit. Original research reproduction (not from DeFiHackLabs). Standalone Foundry PoC, offline anvil_state.json, and full write-up: 2026-04-BlockchainBets_exp in the evm-hack-registry mirror.

Vulnerability classes: vuln/logic/incorrect-calculation · vuln/logic/missing-check · vuln/logic/reward-calculation

Reproduction: the PoC compiles & runs in an isolated Foundry project at this project folder. The fork is served offline from the bundled anvil_state.json (local anvil replays Ethereum state at block 24979555), so no public RPC is required. Full verbose trace: output.txt. BCB token (verified): BlockchainBets.sol. ERC-1155 stake module is unverified on Etherscan (0x3331F635…4f52).

Key info#

TL;DR#

1. Attacker CREATE-deploys a nested factory at nonce 0 that flash-swaps 4_000_000_000 raw BCB (9 decimals) from the UniV2 BCB/WETH pair.
2. Flash funds are split across helper contracts that call the unverified ERC-1155 staking module.
3. Each helper stakes → transforms → unstakes. The transform/unstake accounting inflates redeemable BCB far above the BCB locked in.
4. Inflated BCB is dumped back into the pair; the flash is repaid; leftover WETH is unwrapped to 17.568954659981801955 ETH and forwarded to the attacker EOA.

Background#

Blockchain Bets (BCB) is an ERC-20 (9 decimals) with a Uniswap V2 BCB/WETH pool and an associated ERC-1155 “stake / transform” module at 0x3331…4f52. The module is not verified on Etherscan. On-chain behavior and the attack trace show it:

• Accepts BCB deposits and mints ERC-1155 position ids (stake, selector 0x48c35d46)
• Converts positions across ids (transform, selector 0x9a3c6467)
• Redeems positions back to BCB (unstake / withdraw, selector 0xa2bc66be)
• Tracks per-id supply via ERC-1155Supply-style totalSupply(uint256) (0xbd85b039)

The economic bug is that the redeem path pays out more BCB than was economically deposited for the transformed positions — classic share / accounting inflation on a staking wrapper.

Root cause#

Inflated redeemable balance on stake / transform / unstake#

The vulnerable surface is the unverified ERC-1155 module. From the historical attack bytecode and event log:

So the redeemable BCB claim after transform is not conserved against the BCB escrowed by the module. The surplus is real transferable BCB (token balance increases), which the attacker sells into the UniV2 pool.

Because the module is unverified, the precise Solidity formula is not published; the PoC reproduces the historical CREATE initcode end-to-end so the bug is demonstrated against live pre-attack state without reverse-engineering a full source map.

// Conceptual (RECONSTRUCTED surface — not verified source)
// stake(id0, amount)     → mint ERC1155 id0, pull `amount` BCB
// transform(id0 → id1)   → burn id0, mint id1 with *inflated* units
// unstake(id1, units)    → burn id1, push *inflated* BCB to caller

Why a flash swap is enough#

The pool holds deep enough BCB inventory for a 4e9 raw flash (swap(amount0Out=4e9, …, data)). The inflate step turns that inventory into a much larger BCB balance without a matching BCB lock in the module, so after repaying the flash the attacker still holds a large BCB residual to sell for WETH.

Preconditions#

1. UniV2 pair 0x3beE…4a8A has BCB (token0) and WETH (token1) reserves (pre-attack ~55.28 WETH).
2. ERC-1155 module at 0x3331…4f52 is live and accepts stake/transform/unstake for the chosen ids.
3. Attacker can deploy CREATE initcode that initiates a pair flash swap with a callback into attacker-controlled helpers.

No privileged roles, oracles, or governance actions are required.

Attack walkthrough#

Exact offline PoC numbers from output.txt:

1. CREATE factory at historical address 0xaBB388…f8f4 (attacker nonce 0). Nested helpers are CREATE’d inside the constructor path.
2. Flash swap 4_000_000_000 raw BCB from the pair into the helper tree (UniswapV2Pair.swap → callback).
3. Split + stake/transform/unstake across helpers (0x3ea5…, 0x9a59…, 0x27e75… pattern in the historical tx). ERC-1155 TransferSingle events show stake of 1e9 and unstake of inflated amounts (e.g. 0x06a840bdceb478 raw BCB on one leg).
4. Repay flash by returning BCB to the pair; dump surplus BCB for WETH (pair WETH: 55.2837 → 37.5449).
5. Unwrap WETH → ETH (WETH.withdraw) and forward 17.568954659981801955 ETH to the attacker EOA.

Remediation#

1. Conserve value across stake/transform/unstake: redeemable BCB must not exceed escrowed BCB (or a correctly accrued reward schedule with explicit funding).
2. Invariant tests: property tests that random stake/transform/unstake sequences never increase an account’s claimable BCB above deposits + funded rewards.
3. Verify and publish the ERC-1155 module source; pause stake/transform until the accounting is fixed.
4. Prefer battle-tested vault patterns (ERC-4626-style share math with explicit totalAssets checks) over ad-hoc multi-id transform multipliers.

How to reproduce#

cd /path/to/evm-hack-registry
_shared/run_poc.sh 2026-04-BlockchainBets_exp -vvvvv
# → [PASS] testExploit — Attacker ETH profit: 17.568954659981801955

PoC strategy: replay historical CREATE initcode at fork block 24979555 (see test/BlockchainBets_exp.sol).

Reference: ExVul alert · attack tx 0x879b365b…

Sources & further analysis#

Reproductions & code

• Standalone PoC + full trace: 2026-04-BlockchainBets_exp (evm-hack-registry mirror).
• Attack transaction: view on explorer.

Alerts & third-party analyses

• Original alert / thread: post on X.
• Web3Sec X hacked database: search.
• Rekt leaderboard: search.
• Solodit incident search: search.

These dashboards index community alerts tweets, post-mortems, and independent write-ups. Reach them through the protocol name above to cross-check this reproduction against other analyses.