Skip to content

Instantly share code, notes, and snippets.

@Fitblip
Last active June 6, 2021 05:58
Show Gist options
  • Save Fitblip/3d73c9e2b4f5b3822e9215d16e5cc789 to your computer and use it in GitHub Desktop.
Save Fitblip/3d73c9e2b4f5b3822e9215d16e5cc789 to your computer and use it in GitHub Desktop.

Airdrop XTZ movement cluster analysis

This analysis focuses on XTZ moving around among airdrop recipient accounts. Participants who sent funds to more than 2 addresses who are also on the airdrop recipient list are considered a cluster.

There are a few caveats - namely that the liklihood of including legitimate recipients is higher than the sister analysis that was done to find wallet funding sources. The other is that addresses that had 0 sends to other recipients are excluded from analysis.

Analysis Code + Results

The analysis script can be found here

Running the script produces this output

Then it also prints out cluster data in a summarized view and a view with timestamps per transaction.

This data was then analyzed by hand to look for sends between clusters of common addresses (or if these clusters were self-contained), like Cluster 1, where address tz1ez5hc7wfNKrZ65vVRyC6JBBxZF6UYX3aD recieves funds from tz1RmPNzio2JFSecq8W1bohHsRFxetW3WTsF, tz1WKrHsuJfc3yuJBPskxxFgcTxDcRwBFPtg, and tz1aXJbVA8vUwi7r91ggPhgqFqaqY8isLQBW.

The results from this process can be found here.

Aggregating together all cluster participants you get the following list of addresses to be blacklisted along with a script to tally the new point allocation for the cluster leader and proposed adjustments.

TL;DR These addresses get blacklisted outright and the following allocations are made:

Cluster 1 lead by tz1aXJbVA8vUwi7r91ggPhgqFqaqY8isLQBW now gets 4 point/s {'Quipu_LP_Old', 'Kolibri_LP', 'Dexter_LP', 'Protocol_User'}
Cluster 2 lead by tz1cmanHzitf3UgWgECzLpGUKpzEfmGZaFsw now gets 4 point/s {'Quipu_LP_Old', 'Kolibri_LP', 'Dexter_LP', 'Protocol_User'}
Cluster 3 lead by tz1agwxv1npTpnBZLGYg8kELfjMGjkySdKtC now gets 4 point/s {'Quipu_LP_Old', 'Quipu_LP_New', 'Dexter_LP', 'Protocol_User'}
Cluster 4 lead by tz1NpaSh5FzvsqVCEcLDtfexYyYELz9ys1m7 now gets 3 point/s {'Quipu_LP_New', 'Dexter_LP', 'Protocol_User'}
Cluster 5 lead by tz1aNkb3p9Zcu9FTxyEU9HzAi5fFzSDRfLC3 now gets 4 point/s {'Quipu_LP_Old', 'Quipu_LP_New', 'Dexter_LP', 'Protocol_User'}
Cluster 6 lead by tz1dqTasQcSSEK1UKJG9ifu81VyPnKzdV33T now gets 5 point/s {'Kolibri_LP', 'Dexter_LP', 'Protocol_User', 'Quipu_LP_New', 'Quipu_LP_Old'}
Cluster 7 lead by tz1bpz9S6JyBzMvJ97qPL7TeejkUiLjdkDAm now gets 5 point/s {'Kolibri_LP', 'Dexter_LP', 'Protocol_User', 'Quipu_LP_New', 'Quipu_LP_Old'}
Cluster 8 lead by tz1cSMrgBZXdv63TYyHQb9UBJWueBXYqfA4V now gets 5 point/s {'Kolibri_LP', 'Dexter_LP', 'Protocol_User', 'Quipu_LP_New', 'Quipu_LP_Old'}
Cluster 9 lead by tz1bAnyngBRVCiBueEJoEGoGYo4vPauBXm8L now gets 5 point/s {'Kolibri_LP', 'Dexter_LP', 'Protocol_User', 'Gov_Executor', 'Quipu_LP_Old'}
Cluster 10 lead by tz1bjTaXURxeCucaPiCvRB7F9LouyUqdvn99 now gets 5 point/s {'Kolibri_LP', 'Dexter_LP', 'Protocol_User', 'Quipu_LP_New', 'Quipu_LP_Old'}
Cluster 11 lead by tz1dKuipgWszbTRD8hE5ar2YSDFTAFoy1yRc now gets 4 point/s {'Quipu_LP_Old', 'Quipu_LP_New', 'Kolibri_LP', 'Dexter_LP'}
Cluster 12 lead by tz1bA7aVbV7SpBfr3FExkrTMt8woAdh1zZDJ now gets 3 point/s {'Quipu_LP_Old', 'Quipu_LP_New', 'Dexter_LP'}
Cluster 13 lead by tz1bxpnJwhqrdhf5Dmsp8QusjL9zN4qbDCWs now gets 2 point/s {'Quipu_LP_New', 'Dexter_LP'}
Cluster 14 lead by tz1eM8EAJnBT4uNdGSwVkJh9ABfLVoYQ2mDC now gets 1 point/s {'Quipu_LP_New'}
Cluster 15 lead by tz1e34cyK8Ppa1Sw4WbwMJG6fECUsh9ifbDh now gets 1 point/s {'Quipu_LP_New'}
Cluster 16 lead by tz1dCyFm9jxuhxJEi8UW5sgiS7TFubaNT1J7 now gets 2 point/s {'Kolibri_LP', 'Quipu_LP_New'}

Questions? False Positive?

If you think you were mistakenly included in a cluster and shouldn't have been or have any questions about the analysis, please drop by the discord so we can chat or email us at hello@hover.engineering.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment