arbitrumfoundation / sybil-detection

Arbitrum Sybil Detection

This project aims to remove Sybil addresses from the Arbitrum airdrop, ensuring that only legitimate users receive the airdrop tokens.

---

Methodology

We use on-chain data to identify related addresses owned by the same user and remove entity addresses such as bridges, exchanges, and smart contracts using data from Nansen, Hop, and OffChain Labs. There are also some addresses being removed by manual inspection, such as donation addresses, l2 alias, etc.

With those cleaned up, 2 graphs are generated for this study:

Graph 1: In this graph, each transaction with msg.value is treated as an edge with their (from_address, to_address)

Graph 2: In this graph, each funder/sweep transaction is treated as an edge with their (from_address, to_address)

• funder transaction is the first ether transfer to an account
• sweep transaction is the last ether transfer from an account

---

Sybil Cluster Identification

Clusters are generated by partitioning the above graphs into strongly connected and weakly connected subgraphs. Large subgraphs are broken down using the Louvain Community Detection Algorithm, providing more refined results and eliminating Sybil addresses more accurately.

We identify Sybil clusters based on known patterns, here are some examples

• Addresses transferring funds in a cluster of more than 20 addresses
• Addresses that are funded from the same source
• Addresses with similar activity

---

Examples

Cluster 319 with 110 eligible addresses

Sample address: 0x1ddbf60792aac896aed180eaa6810fccd7839ada

Cluster 1544 with 56 eligible addresses

Sample address: 0xc7bb9b943fd2a04f651cc513c17eb5671b90912d

Cluster 2554 with 121 eligible addresses

Sample address: 0x3fb4c01b5ceecf307010f84c9a858aeaeab0b9fa

Cluster 3316 with 65 eligible addresses

Sample address: 0x15bc18bb8c378c94c04795d72621957497130400

---

Inputs Used

1. Raw Eligibility List (from Nansen)

2. Excluded Entities (from Nansen)

3. CEX Deposit Addresses (from Nansen)

4. CEX Deposit Addresses (traced from CEXs hot wallets)

5. Unique transaction and traces (from,to) Arbitrum

6. Unique transaction and traces (from,to) Ethereum

7. Internal Address list from OffChain Labs

8. Hop Blacklist

9. Hop eliminatedSybilAttackers

10. Nansen address tags

11. Other active addresses tagged manually

sybil-detection's Issues

Inquiry about code release

Hi, I am a student conducting research related to sybil detection.
Can u share me your code? It would helpful for my research~ thx.

148,595 Sybil airdrop addresses

The biggest party would receive 1.83 million tokens alone.

Further details:
https://twitter.com/x_explore_eth/status/1638452183682416641
https://mirror.xyz/x-explore.eth/AFroG11e24I6S1oDvTitNdQSDh8lN5bz9VZAink8lZ4

All credit to x_explore.eth

arb/sybil-det3ction

Release the code please

Hello, I work on Sybil detection at Pometry and with Gitcoin.

I like your approach and would like to integrate your workflow with my own to create the best possible sybil detection code.

Would you mind sharing the code and data?

Sybil attacker will take 3026125 token ARB if the team do nothing!

Hello, I found something interesting about this wallet address: 0x59d4087f3ff91da6a492b596cbde7140c34afb19

He made 2,417 transactions within 6 hours, that includes sending small ETH to 1656 different wallets :

Details: https://arbiscan.io/txs?a=0x59d4087f3ff91da6a492b596cbde7140c34afb19

And each recipient wallets, he made 2 interactions with ARB token contract:
For example,

1. 0x3E5A2B1020c454079f5A7702fa204752C584d6A0
   
   Tx: https://arbiscan.io/address/0x3e5a2b1020c454079f5a7702fa204752c584d6a0

2. 0x3aBeC2bbEc31c978a4a7e5b0cD2090cB759A0c01
   
   Tx: https://arbiscan.io/address/0x3abec2bbec31c978a4a7e5b0cd2090cb759a0c01

I check all 1656 recipient wallets above:https://github.com/stanlagermin/sybil-wallet-list/blob/main/sybil_wallets.csv
with: https://cointool.app/airdrop/arb

IT MAKES ME SHOCK!!!!!! 2800875 ARB TOKEN

I also recognize that every recipient wallet has at least one transaction relate to this wallet: 0xcc577C130c019529FF1e721F9BEeA24a7DC1402D

For example:

1. 0x3E5A2B1020c454079f5A7702fa204752C584d6A0
   
   Tx: https://arbiscan.io/tx/0x9556ae9962c8034eb98f3f817eb5ecbb6d3e588fb71c70b84dcc5247dfcda998

2. 0x3aBeC2bbEc31c978a4a7e5b0cD2090cB759A0c01
   
   Tx: https://arbiscan.io/tx/0x73e7f173ced28ab9aecf019d050609f2a85367917fce3bef56aa9b37f23d8fe7

From my point of view, that guy did all actions above is an airdrop farmer or a hacker because some people say that their wallet got hacked and being in a Sweeper-bot. Anyway, the team should do something to prevent getting at least 2800875 ARB Token from bad person.

Raw Data

Could you please share your raw data for sybil detection?

Link to HOP blacklist is not working

https://raw.githubusercontent.com/hop-protocol/hop-airdrop/master/src/data/c.csv

for me returns a 404

this comes from the ReadMe

also fwiw - this kind of analysis is conducted by the OpenData Community regularly, initially as a part of defending Gitcoin rounds.

thank you for publishing information on your approach

Launch a sybil finding bounty

Hi admins. Will you launch a sybil finding bounty like Hop Protocol did? I think there are so many people who want to participate

Incorrect Sybil account

According to these criteria some people seems to have funded from kucoin or Binance and are now considered as sybil
Isn't that an overdo?

3,002 Provable Sybilooors w/ 5+ Accounts Each to Receive 7,847,875 Tokens (~$10m currently)

I conducted an analysis of addresses on the ARB airdrop list and found around $10m worth of sybil accounts.

The data is 100% provable unlike the prior sybil analyses with many false positives. Will share more if anyone cares.