The @wormholecrypto Sybil filtering criteria was just released today

It’s important to always study filter criteria in depth to understand the thought process behind it and how you can use this for your future farming methods

Ex. If you read the Arbitrum criteria you would of known that they gave a -1 point for having less than 0.005 ETH on your wallet

If you took this lesson and applied it to Starknet then you would of not got filtered out

Lets go over the criteria in detail and see what the biggest takeaway are that you can use to improve your future farming

1/ Ownership graph

Wormhole contributors constructed an ownership graph that assigns a unique user ID to a cluster of wallets that belong to the user

Wallets which interacted with any one of Wormhole’s ecosystem applications were included

The ownership graph consolidates all transactions a user may have conducted across different ecosystems under a single user ID

This ensures that a user is evaluated based on all of their transactions, rather than on fragmented views of their activities in any particular ecosystem

This user ID method of determining a “family” of connected wallets is similar to the typical wallet cluster labeling methods used to identify wallets highly connected and correlated with each other based on their onchain history

**Takeaway: **organizing and managing your wallets into small clusters is crucial

2/ Source of Funds Analysis

The analysis revealed two types of source funding attacks:

Diffusion Funding: Wallet A funds Wallets B, C, D ….

Sequential Diffusion Funding: Wallet A funds B, Wallet B funds Wallet C…

**Takeaway: **wallet funding methods matter don't be lazy, if your funding everything from the same origional source its going to look obvious

"CEX is best" still stands, withdrawing randomized amount at random time from multiple different CEXs is and will always continue to be the meta

3/ Behavioral Clustering Analysis

The algorithm maps transactions to unique IDs based on timing, cadence, and action, to match identical or nearly identical behaviors

This separates high density large scale automated wallet actions from low density genuine user actions

Wormhole contributors used this identification strategy only in conjunction with other sybil detection mechanisms

**Takeaway: **Be organic. Always think from the perspective of "what would an organic users do?"

If you fund a large cluster of wallets all from the same CEX, with the same asset, on the same chain, with nearly (or a very close) the same exact withdrawal time stamp....

This probably is NOT organic compared to randomizing and adding variation to all of these variables

4/ Transaction Spam Analysis

Another common large scale action involved rapid sequential transactions of identical token amounts, either between two chains or across multiple chains

Wallets with transactions that both matched this activity at **abnormally high **frequencies and coincided with the large influx of sybil activity starting in December 2023 may have been tagged as indicative of spam-like behavior and subsequently removed from the airdrop eligibility dataset

**Takeaway: **Those Arzidor L0 threads back in the day telling you you'll qualify for L0 by spamming 50x $0.02 txns within 1 hr .... prolly ngmi anon

Although we don't have Wormhole's definition for "abnormally high frequencies"

Just don't spam a gorillion small volume txns between a few wallets as fast as you can

And I can already hear the reply guys typing now saying how/why merkly caused them the to be ineligible

No, if that was the only interaction you did then yes prolly ngmi, but if you had even a few hundred in volume over 1-2+ txn you were fine, I used merkly to refuel and mint and bridge NFTs on all my wallets and they qualified

Now with all of this in mind imagine if you were able to hit qualifying criteria, avoid Sybil detection, and save yourself a significant amount of time by automating all of these actions across multiple wallets?

That's exactly what I'm doing with Zenith_farms

(sorry, not sorry I'm shilling Zenith again and will continue to do so)

But I'll explain why exactly your able to avoid sybil detection even when using an automated tool and why I think it's +EV to "hedge" by having some sort of automation allocation in your airdrop farming portfolio

1/ Ownership graph

Avoided by allowing you to organize your wallet groups effectively in small (5 or less ideally) cluster sizes in order to avoid sybil detection

Paired with the ability to highly randomize the interaction types and the time delay between those interactions makes them highly distinct from each other and virtually indistinguishable from an organic user

2/ Source of Funds Analysis

You can avoid this funding trap completely by having each wallet individually funded from a CEX address with variability in the asset type, quantity, and time funds are withdrawn.

(This can also be done automatically using exchange API keys)

This completely removes the need to use Diffusion and Sequential Diffusion funding methods in order to fund all farming wallets

3/ Behavioral Clustering Analysis

Similar to the point above using Zenith to not only withdraw from multiple different CEXs but also to randomize the amount withdrawn

+ having a time delay between withdrawals ensures a unique timestamp and withdrawal fingerprint when funding wallets = no behavioral clustering

4/ Transaction Spam Analysis

Zenith avoids these sybil like patterns by being able to craft individualized and randomized interaction routes between bridges and various DApps

With also having time delays between interactions this removes the need to send a high frequency of transactions between various wallets in order to build out a wallet interaction or transaction count

You can just have each wallets que to tasks to do their own thing with as long of a time delay as you want

I fully advocate the "Manual Chad/Grinder" farming method as always

But I also think its beneficial to allocate some of your farming effort to the automation (dark/ not as sexy) side of farming as well

Trust me once you get the hang of it will feel painful to going back to manual only farming/grinding

If you want to try out the tool you can start with just a small amount of wallets (only costs 0.02-0.025 ETH for 5 wallets/month) and there are several tutorial videos that I've made to help you get started (in Zenith discord)

You can get onboarded by using the code “ insightful “ to get 5% off when you purchase a module

https://app.zenith.farm/?referral=Insightful

You can also watch this short playlist of tutorial videos to understand how the tool works

Disclosure - I have a vested interest that Zenith does well, I’m not on the team but I’m involved in helping make content and doing research and help guide the ecosystem funds, and have exposure to the ecosystem token as well (but you don’t need the token to use the tool)

The @wormholecrypto Sybil filtering criteria was just released today

It’s important to always study filter criteria in depth to understand the thought process behind it and how you can use this for your future farming methods

Ex. If you read the Arbitrum criteria you would of known that they gave a -1 point for having less than 0.005 ETH on your wallet

If you took this lesson and applied it to Starknet then you would of not got filtered out

Lets go over the criteria in detail and see what the biggest takeaway are that you can use to improve your future farming

1/ Ownership graph

Wormhole contributors constructed an ownership graph that assigns a unique user ID to a cluster of wallets that belong to the user

Wallets which interacted with any one of Wormhole’s ecosystem applications were included

The ownership graph consolidates all transactions a user may have conducted across different ecosystems under a single user ID

This ensures that a user is evaluated based on all of their transactions, rather than on fragmented views of their activities in any particular ecosystem

This user ID method of determining a “family” of connected wallets is similar to the typical wallet cluster labeling methods used to identify wallets highly connected and correlated with each other based on their onchain history

**Takeaway: **organizing and managing your wallets into small clusters is crucial

2/ Source of Funds Analysis

The analysis revealed two types of source funding attacks:

Diffusion Funding: Wallet A funds Wallets B, C, D ….

Sequential Diffusion Funding: Wallet A funds B, Wallet B funds Wallet C…

**Takeaway: **wallet funding methods matter don't be lazy, if your funding everything from the same origional source its going to look obvious

"CEX is best" still stands, withdrawing randomized amount at random time from multiple different CEXs is and will always continue to be the meta

3/ Behavioral Clustering Analysis

The algorithm maps transactions to unique IDs based on timing, cadence, and action, to match identical or nearly identical behaviors

This separates high density large scale automated wallet actions from low density genuine user actions

Wormhole contributors used this identification strategy only in conjunction with other sybil detection mechanisms

**Takeaway: **Be organic. Always think from the perspective of "what would an organic users do?"

If you fund a large cluster of wallets all from the same CEX, with the same asset, on the same chain, with nearly (or a very close) the same exact withdrawal time stamp....

This probably is NOT organic compared to randomizing and adding variation to all of these variables

4/ Transaction Spam Analysis

Another common large scale action involved rapid sequential transactions of identical token amounts, either between two chains or across multiple chains

Wallets with transactions that both matched this activity at **abnormally high **frequencies and coincided with the large influx of sybil activity starting in December 2023 may have been tagged as indicative of spam-like behavior and subsequently removed from the airdrop eligibility dataset

**Takeaway: **Those Arzidor L0 threads back in the day telling you you'll qualify for L0 by spamming 50x $0.02 txns within 1 hr .... prolly ngmi anon

Although we don't have Wormhole's definition for "abnormally high frequencies"

Just don't spam a gorillion small volume txns between a few wallets as fast as you can

And I can already hear the reply guys typing now saying how/why merkly caused them the to be ineligible

No, if that was the only interaction you did then yes prolly ngmi, but if you had even a few hundred in volume over 1-2+ txn you were fine, I used merkly to refuel and mint and bridge NFTs on all my wallets and they qualified

Now with all of this in mind imagine if you were able to hit qualifying criteria, avoid Sybil detection, and save yourself a significant amount of time by automating all of these actions across multiple wallets?

That's exactly what I'm doing with Zenith_farms

(sorry, not sorry I'm shilling Zenith again and will continue to do so)

But I'll explain why exactly your able to avoid sybil detection even when using an automated tool and why I think it's +EV to "hedge" by having some sort of automation allocation in your airdrop farming portfolio

1/ Ownership graph

Avoided by allowing you to organize your wallet groups effectively in small (5 or less ideally) cluster sizes in order to avoid sybil detection

Paired with the ability to highly randomize the interaction types and the time delay between those interactions makes them highly distinct from each other and virtually indistinguishable from an organic user

2/ Source of Funds Analysis

You can avoid this funding trap completely by having each wallet individually funded from a CEX address with variability in the asset type, quantity, and time funds are withdrawn.

(This can also be done automatically using exchange API keys)

This completely removes the need to use Diffusion and Sequential Diffusion funding methods in order to fund all farming wallets

3/ Behavioral Clustering Analysis

Similar to the point above using Zenith to not only withdraw from multiple different CEXs but also to randomize the amount withdrawn

+ having a time delay between withdrawals ensures a unique timestamp and withdrawal fingerprint when funding wallets = no behavioral clustering

4/ Transaction Spam Analysis

Zenith avoids these sybil like patterns by being able to craft individualized and randomized interaction routes between bridges and various DApps

With also having time delays between interactions this removes the need to send a high frequency of transactions between various wallets in order to build out a wallet interaction or transaction count

You can just have each wallets que to tasks to do their own thing with as long of a time delay as you want

I fully advocate the "Manual Chad/Grinder" farming method as always

But I also think its beneficial to allocate some of your farming effort to the automation (dark/ not as sexy) side of farming as well

Trust me once you get the hang of it will feel painful to going back to manual only farming/grinding

If you want to try out the tool you can start with just a small amount of wallets (only costs 0.02-0.025 ETH for 5 wallets/month) and there are several tutorial videos that I've made to help you get started (in Zenith discord)

You can get onboarded by using the code “ insightful “ to get 5% off when you purchase a module

https://app.zenith.farm/?referral=Insightful

You can also watch this short playlist of tutorial videos to understand how the tool works

Disclosure - I have a vested interest that Zenith does well, I’m not on the team but I’m involved in helping make content and doing research and help guide the ecosystem funds, and have exposure to the ecosystem token as well (but you don’t need the token to use the tool)