whodl
Solana holder intelligence

Who’s really holding?

Follow the wallets, not the noise. Drop a token’s mint on X, Telegram, Discord or right here. Whodl maps the biggest wallets, flags bundled and linked holders, and shows who added, who sold and where the money went. Every number links to its transaction.

  • 40max wallets per scan
  • 24h/7dposition windows
  • 0wallets to connect
whodl://reply X

> @WhodlBot scan 2j5SaS…nW2ZM

$CLAUDIA · 28 large accounts · 24h + 7d

MIXED New large positions opened while several holders cut exposure.

Top seller dumped 3.82M, bought back 1.69M. Net exit: 2.13M.

LINKED Check the bubble map for wallets funded from one source.

Full report $CLAUDIA holder report
$CLAUDIA example · Oct 17, 202628 accounts examined+10.50M added in 24hGpGJDG −74.1% in 24h18.23M position → 0 in 7d1.69M bought backRef price $0.009049Verified CARDS → CLAUDIA swaps
Use it anywhere

Scan on the web.
Or in your chat.

Paste a mint below, add the bot to your Telegram group or Discord server, or tag it on X. Every channel produces the same report.

WEB

Use the exact contract address. You can also ask by token name in web chat; ambiguous names need a contract. Or open the web chat.

Add to TelegramComing soon Add to DiscordComing soon
Tag @WhodlBot on XConnection pending — web scans are live
Bubble map · bundle detection

Ten wallets.
One hand?

Every report draws the top holders as bubbles sized by their share of supply, then connects the ones that look coordinated. Linked supply and bundled supply are counted up front, with a receipt for every link.

  • Same funderFirst funded by the same wallet, or one holder funded another.
  • BundleTokens received in the same transaction, or first buys in the same block.
  • Direct transfersTokens sent straight from one holder to another.

A shared funder is a signal, not proof. Exchanges fund unrelated wallets all the time, and the report says so.

Linked cluster 6.9% Bundle 3.2% Unlinked
Illustration. Bubble sizes from the CLAUDIA snapshot; links are examples.
How it works

One mint in.
One report out.

The arithmetic happens in code against finalized chain state. AI only explains what the numbers already support, and a second, independent pass checks every claim.

  1. 01

    Send a mint

    Paste the exact contract address on the site, in Telegram or Discord, or in an X mention. Ambiguous tickers get a question, not a guess.

  2. 02

    Snapshot the holders

    Up to 40 large token accounts, their funding sources and their balances rebuilt for the last 24 hours and 7 days.

  3. 03

    Calculate, then audit

    Net changes, clusters and supply share come from integer math. A writer drafts the brief and a separate auditor checks it.

  4. 04

    Reply with a link

    A short, grounded reply where you asked, with a permanent report URL. Repeat requests within five minutes share one report.

Why net position

Gross sales lie.
Net doesn’t.

A wallet that dumped 3.8 million tokens looks like it ran for the exit. If it also bought 1.7 million back, the real reduction is smaller, and that changes the read.

Whodl puts gross sales, buybacks and the position that remains side by side, in both windows, so you can see which moves actually changed exposure.

GpGJDG…JYhT$CLAUDIA · 24h
Starting position2,872,083
Gross sold−3,816,371
Bought back+1,689,452
Net sold−2,126,919
Remaining position745,164
Position down 74.1%, not the 133% the gross sales alone suggest. Tokens, from the CLAUDIA example.
Inside the report

The full story
behind every reply.

Every report uses the same layout, is stored permanently and downloads as JSON. A refresh creates a new report instead of rewriting the old one.

Takeaway

The read, in two lines

Encouraging and concerning evidence, each linked to the rows behind it.

Bubble map

Who’s connected

Linked and bundled wallets, the reason for each link, and their combined supply.

Position changes

The moves that matter

Largest balance changes, switchable between 24 hours and 7 days.

Sellers & buybacks

Who actually left

Gross sold, bought back and net sold per seller, in tokens, supply share and USD.

Rotations

Where sellers went

Assets received in the same verified swap, like SOL, USDC or another token.

Sources & coverage

What was and wasn’t seen

Snapshot time, price time, history coverage and anything marked unknown.

Evidence boundaries

What a report
won’t claim.

  • Accounts aren’t people. One owner can hold several token accounts, and exchange or pool custody is only partly labeled.
  • Unknown is never zero. Missing or truncated history leaves a change unknown instead of counting it as no change.
  • Only verified swaps count as trades. A Jupiter router call, the owner’s signature and one clear opposing asset. Everything else is a transfer.
  • USD is a reference. Dollar figures use one timestamped price, not what a seller actually received.
In your chats

A bot that only
speaks when asked.

  • Replies only when asked. No unsolicited messages, quote posts or promotion.
  • One reply per request. Uncertain sends go to human review, never an automatic retry.
  • Easy opt-out. Tell it “stop” and it won’t reply to you again.
  • No wallet, no token. Nothing to connect, sign or buy to read a report.
See it on real data

See a real report
before you scan.

The CLAUDIA example covers 28 large accounts across 24 hours and 7 days, with every seller, buyback and rotation traced to its transaction.

System status
Checking service status…
Web scans use live on-chain data. Messaging channels become available as their account connections are completed.