Skip to content

Instantly share code, notes, and snippets.

@nikkaroraa
Last active February 24, 2026 12:14
Show Gist options
  • Select an option

  • Save nikkaroraa/c61b496738c7629f4f2d3580c1a0f590 to your computer and use it in GitHub Desktop.

Select an option

Save nikkaroraa/c61b496738c7629f4f2d3580c1a0f590 to your computer and use it in GitHub Desktop.

Sybil Analysis - Questing Season 1

Data snapshot: Feb 24, 2026. Type tracking deployed Feb 18.


Summary

24,902 wallets exhibit identical behavior: bridge(s) of ~$18K, complete all one-time quests, skip Connect X and Weekly TVL, hold $0 in vaults. Total volume attributed: $468.8M.


Detection Method

Sybil wallets are identified by bridge amount:

  • Any user with any bridge to Katana valued between $17,000 and $19,000
SELECT DISTINCT "userId" FROM "Intent"
WHERE type = 'BRIDGE_TO_KATANA'
  AND "usdValue" BETWEEN 17000 AND 19000

The amount range is consistent with bridging ~10 ETH per wallet (at ETH ~$1,840 during Feb 18-24). Most sybils bridge exactly once, but 547 users bridged multiple times in this range — an earlier filter requiring exactly one bridge missed these multi-bridge sybils.

Scale

Metric Sybil Organic Total
Users 24,902 (39.3%) 38,468 (60.7%) 63,370
Bridge volume $468.8M (>99%) $235K (<1%) $469.0M
Bridge intents 25,514 (94.2%) 1,572 (5.8%) 27,086
Swap intents ~24,184 (84%) 4,488 (16%) ~28,672
Swap volume ~$4.7M (86%) $760K (14%) ~$5.5M
Vault deposit intents ~23,263 (93%) 1,715 (7%) ~24,978
Vault deposit volume ~$45K (2%) $2.0M (98%) ~$2.1M
TVL held $0 (0%) $20.88M (100%) $20.88M

Organic numbers are from fresh queries. Sybil swap/vault counts are derived (total minus organic) and approximate.

Bridge Value Distribution

Picture 27,000 people at an ATM. If they were real people, they'd withdraw $20, $500, $3,000, all different amounts. Our data shows 25,000 of them withdrawing between $18,252 and $18,505. That's a $253 window. That does not happen organically.

Histogram

Amount Range Bridges % of Total Volume
$0 (dust) 422 1.6% $126
$1-$100 1,026 3.8% $16K
$100-$1K 49 0.2% $16K
$1K-$5K 17 0.1% $36K
$15K-$18K 825 3.1% $14.5M
$18K-$20K 24,669 91.3% $453.9M
$50K-$100K 2 0.0% $167K

91.3% of all bridge intents are $18K-$20K. The amount is consistent with bridging exactly ~10 ETH per wallet (at ETH price ~$1,840 during Feb 18-24, 10 ETH = ~$18,400).

Fine-grained breakdown ($17K-$19K)

Range Bridges
$17,000-$17,100 14
$17,100-$17,200 20
$17,200-$17,300 15
$17,300-$17,400 16
$17,400-$17,500 269
$17,500-$17,600 169
$17,600-$17,700 115
$17,700-$17,800 13
$17,800-$17,900 30
$17,900-$18,000 174
$18,000-$18,100 651
$18,100-$18,200 1,667
$18,200-$18,300 4,525
$18,300-$18,400 4,372
$18,400-$18,500 6,119
$18,500-$18,600 5,665
$18,600-$18,700 1,616
$18,700-$18,800 54

The core cluster is $18,100-$18,700 (23,964 intents). The variation is ETH price movement during the week.

Bridges Per User

Bridges per user Users % of Users Avg Volume/User
1 25,220 96.9% $17,738
2 687 2.6% $25,580
3 98 0.4% $31,042
4+ 28 0.1% varies

96.9% of bridge users bridge exactly once. The remaining 3.1% (813 users) bridge multiple times — 547 of these have at least one bridge in the $17K-$19K range and are included in the sybil set. The multi-bridge sybils are likely the same operator running wallets through multiple rounds.

Random Sample of Bridge Intents

20 randomly selected bridge intents - 17 out of 20 fall in the $18,178-$18,678 range:

Intent ID User Date USD Value
0x4b9e...a701 0xe4a3...4c2e Feb 18 $18,178
0x2148...1301 0x61eb...8464 Feb 19 $18,256
0xbff1...a601 0xf541...1808 Feb 22 $11
0x98f9...8c01 0x39a0...38cd Feb 21 $18,533
0xd000...7a01 0x3d62...eefa Feb 19 $18,477
0x4f53...8401 0xbc03...d8ba Feb 21 $18,673
0xf994...3401 0xe160...820c Feb 19 $18,400
0x436c...5e01 0xcd2d...9934 Feb 21 $18,678
0xb9cf...0001 0x6efb...4817 Feb 21 $18,388
0x1bea...1f01 0xd44c...ea47 Feb 19 $18,463
0x7b8e...f901 0x8d71...b82e Feb 19 $18,485
0x30a6...1501 0xa2bc...2d40 Feb 20 $18,359
0x9f7d...5a01 0x44c8...f054 Feb 19 $18,271
0x9985...af01 0xa106...49b6 Feb 19 $18,441
0xed55...2e01 0x2d73...2b74 Feb 19 $18,485
0x0ae5...5c01 0x4f6c...5820 Feb 20 $18,433
0x5c85...d201 0x4c57...77c7 Feb 22 $18,203
0xa681...bf01 0x35c9...0de1 Feb 20 $18,247
0x2e98...9a01 0xd8b9...04b8 Feb 20 $18,421
0x3094...e301 0xcd58...2d9fe Feb 22 $18,507

Quest Completion - Sybil vs Organic

Quest Raw Organic Sybil Sybil %
Verify Wallet ~63,000 38,002 ~25,000 ~40%
Fund Wallet ~52,000 26,975 ~25,000 ~48%
First Swap ~49,000 25,024 ~24,000 ~49%
Swap Major Asset ~49,000 24,973 ~24,000 ~49%
First Vault Deposit ~48,400 25,423 ~23,000 ~47%
Weekly Swap ~53,000 29,197 ~24,000 ~45%
Connect X 7,068 7,068 0 0%
Weekly TVL 4,753 4,753 0 0%

Organic numbers are from fresh queries. Raw/sybil are approximate (raw has grown slightly since last full count).

Connect X and Weekly TVL are the only quests sybil wallets don't complete. This makes sense:

  • Connect X requires a real X account (cost/risk too high to create fake accounts at scale)
  • Weekly TVL requires holding $50+ in vaults for a week (ties up capital)

The Playbook

The sybil operator runs ~25K wallets through this exact sequence:

  1. Create fresh wallet
  2. Bridge 10 ETH ($18.4K) — sometimes multiple rounds from the same wallet
  3. Complete fund_wallet quest
  4. Do a swap (completing first_swap, swap_major_asset)
  5. Do a dust vault deposit (completing first_vault_deposit)
  6. Do a weekly swap (completing weekly_swap)
  7. Skip Connect X (0%) and Weekly TVL (0%)
  8. Move capital out — sybils hold $0 in vaults

Impact on Swap Gaming Analysis

The sybil wallets distort the swap gaming numbers significantly:

Raw (all users)

Bucket Swaps %
Below $10 min ~800 2.8%
Bare minimum ($9.50-$11) ~460 1.6%
$11-$100 ~1,400 4.9%
$100-$500 ~25,700 89.6%
$500+ ~300 1.0%

Organic only

Bucket Swaps %
Below $10 min 1,373 30.6%
Bare minimum ($9.50-$11) 1,404 31.3%
$11-$100 1,382 30.8%
$100-$500 287 6.4%
$500+ 42 0.9%

The raw data made swaps look organic (90% at $100-$500). The organic-only data shows 62% are at or below the $10 minimum, with only 6.4% in the $100-$500 range. The sybil wallets were doing uniform ~$196 swaps, masking the real behavior. The $100-$500 bucket collapsed from 23% to 6.4% when the corrected filter removed multi-bridge sybils that the original filter missed.

Daily Bridge Activity ($18K-$19K range)

Shows whether sybil activity is ongoing or has stopped. Run this to get the latest:

SELECT
  DATE(i.timestamp) AS day,
  COUNT(*) AS bridges_18k_19k,
  COUNT(DISTINCT i."userId") AS unique_users,
  ROUND(SUM(i."usdValue")::numeric, 2) AS total_volume
FROM "Intent" i
WHERE i.type = 'BRIDGE_TO_KATANA'
  AND i."usdValue" BETWEEN 18000 AND 19000
GROUP BY DATE(i.timestamp)
ORDER BY day;

Top Users by Total Bridge Volume

User Bridges Total Volume Avg/Bridge First Last
0xcb9b...1086 6 $110,906 $18,484 Feb 18 Feb 19
0x4f0a...181f 2 $100,089 $50,045 Feb 18 Feb 18
0xe359...c65 4 $74,139 $18,535 Feb 21 Feb 22
0x7b9e...68ac 4 $74,017 $18,504 Feb 19 Feb 19
0xa699...09d 4 $73,979 $18,495 Feb 19 Feb 19
0x43e9...8268 4 $73,795 $18,449 Feb 18 Feb 19
0x948b...50d6c 4 $70,739 $17,685 Feb 22 Feb 23
0x0e79...5f2e2 4 $70,734 $17,683 Feb 23 Feb 23
0x0a89...ed184 4 $70,631 $17,658 Feb 23 Feb 23
0x7d94...f50a 4 $69,981 $17,495 Feb 23 Feb 23

The top user (0xcb9b) made 6 bridges averaging $18.5K each - likely the operator testing. Users with 4 bridges each at ~$18.5K are likely the same operator with wallets that did multiple rounds.

User #2 (0x4f0a) is an outlier - 2 bridges averaging $50K. This one may be legitimate.

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