zkSync - Guild Community Data Analysis

[Jkd-eth]

User Story:

As a bxPana, Community lead at zkSync I want to understand the make up and distribution of the Sybils within the Guild role holders So that I can make further community decisions

Acceptance Criteria

GIVEN all holders of the zkSync Guild role WHEN we pull the data and analyze for each of the users THEN we can make determinations about the quality of the users

Pull the addresses for all holders of the guild role

We need to understand and answer the following questions (similar to the Scroll & etherFi analysis)

• What is the make up of known Sybils?
• How many Sybil rings and what size are the rings?
• Suspicious addresses (low zkSync activity <10 transactions and 0 eth activity)
• Eth model score for those with > n transactions on Eth
• How does the makeup vary between level 1 (15 score) and level 2 (20 score)

Product & Design Links:

#### Tech Details: #### Open Questions: - [ ] What do we want to get from Guild to help us with the data analysis? #### Notes/Assumptions:

[stefi-says]

• Guild- zksync address got, now Axal flow is getting the address. The is on hold till the data gathering ends. Hopefully Monday I can start the day working on it already

[stefi-says]

steps: [x] get the address that qualify for level 1 and 2 - getting not matching number of users compared with the UI, Checking to see if we can get the lists with Guild

[x] run the rings script - On going
[x] analysis the number of eth transactions
[x] analyse the first funder been part of the same list - code ok
[x ] join the flags for this behavior and flags for the suspicious behavior thresholds :

• have only 10 zksync transaction and no eth activity
• have only 10 zksync transaction and eth activity
• have less then 15 ZKsync trx, have less then 20 Zksync trx

[x] get this proportions for level 1 and level 2 Guid members.

[stefi-says]

Results available on this document . Im was not eble to get valuable results when running the eth model on the sample data. I was getting features with 0 as value . Will create a card to verify what could be the problem. @Jeremy-Gitcoin , Iwould like to know if the document is enough or you are missing information to share with them.

[stefi-says]

Given teh high number of sybils, @Jeremy-Gitcoin asked for a deep dive on:

• [x] code, check everything is correct on the calulcaitons and analysis minor issue on the sibil ring script , already fixed. The script is runnig to check the changes on proportion of sybil rings but we do not expect significant changes

• [x] sybils stamps , what is the combination of stamps this sybils are using to pass Paasport?

• [ ] @NadjibBenlaldj how this sybils influence or next re scoring of stamps?

[stefi-says]

Hey guys, here is a spreadsheet containing the sybils found by sybil ring , their credentials and a tab containing the distribution of these credentials. ZkSync Guild Sybil Ring Credentials @Jeremy-Gitcoin @NadjibBenlaldj let me know if you have any other doubt

[NadjibBenlaldj]

@stefi-says Could you please add to the sheet ring_id + level. And if you already have this data, credentials for non sybil addresses ?

[stefi-says]

hey added it here guild zksync address with credentials and levels