Sybil Daddy

Sybil Daddy filters out bad actors in the airdrop process using mainly 3 ways: 1.) We go through the airdrop eligible addresses and use their social graphs=> engagements on their profiles, recasts, likes & also the quality of their followers to rank them ( powered by Karma3Labs ) 2.) For each user we analyze their content using a classification AI model to determine how many of them are probably bots and how many of them are human generated. This also attaches a score to each airdrop address 3.) For Each User we go through their onchain data to see the quality of their transactions and determine their score

We aggregate these scores and multiply it with their airdrop eligibility amount, which means that bots and malicious actors get a negative factor while the good actors get theirs boosted.

We use @Pinata API's to fetch the latest casts from users who we want to analyze and feed that content to a classification model to get a score based on how much of it is AI/human generated which helps us in determining a negative/positive score for them.

We deployed our Airdrop Factory and Airdrop Main contracts on @Base. The frames talk to these contracts so that users can claim their airdrop.

We used @frames.js to develop frames in the app.

We used @Airstack to fetch Farcaster data for a particular user and use it to determine things like their FID, profile details etc.

@Dynamic is used as the wallet connector for the app.