sol mainnet · slot 448,213,907
Sign inConnect wallet
§ 03.3 Product

Analyse a Solana wallet you have proven you control.

Win rate, net PnL, closed and open counts and top tokens across your plan's window, for wallets you have attached and no others.

One trade tells you almost nothing. The token ran or it did not, and your fill was somewhere on that candle. Read sixty of them together over ninety days and the subject changes: it stops being about tokens and starts being about you. Wallet research is that read. The same dossier, run over a wallet instead of a single position.

It works on wallets you have attached to your account. Attached means you proved you control the address. You cannot paste a whale address off a leaderboard and get a report back, because research on addresses you do not control is not live. If that is what you came here for, this page is not going to imply otherwise.

Attached means you signed for it

Attaching a wallet is one off-chain message signature. Never a transaction, so nothing moves and nothing gets approved. What it does is prove the address is yours. After that, Dossier reads every closed swap on it and the aggregate view has something to compute from.

You can start with an email and attach a wallet later. Until one is attached there is nothing to read, so the report stays empty rather than being estimated from nothing.

How many wallets you hold at once is a plan limit: one on Free, three on Pro, ten on Degen. Delete a wallet and its trades, positions and analyses are hard-deleted immediately. The aggregate numbers go with them.

What the report gives you

Run it and you get the wallet reduced to a handful of figures, bounded by your plan window.

Each figure is computed from your on-chain history. The model does not produce numbers anywhere in Dossier. Where the AI writes, it is interpreting figures it was handed.

  • Win rate: the share of closed positions in the window that finished green.
  • Net PnL: the sum across those closed positions.
  • Closed count: how many positions actually resolved.
  • Open count: how many are still running.
  • Top tokens: the names that dominate the window.

Win rate and net PnL only mean something together

A 70% win rate sitting next to a negative net PnL is a specific shape with specific arithmetic behind it. The green trades were closed small. The red ones were held long. None of that is visible in any single trade, because each one looked fine on its own.

In the graded journal that shape has names. EXIT_TOO_EARLY and BAG_HELD, tagged trade by trade, flagged from the second occurrence with a running count. The wallet view is where you see what they add up to across the window. The taxonomy tells you what you do. The aggregate tells you the size of it.

The inverse shape is worth as much. A 30% win rate next to a positive net PnL reads badly and is not the leak. You can spend months grinding at a hit rate that was never the problem.

The open count is the half that has not resolved

Closed and open counts are not decoration. Forty closed and twenty-five open over ninety days means twenty-five positions have not settled anything yet, and the net PnL figure only covers the forty that did. A wallet can read flat because the rest of it has not been realised.

Dossier does not tell you what to do about that. It does not forecast whether any of them recover, and it will never suggest selling one. The Live tab tracks open positions with live prices and RSI, and flags when you are repeating a pattern that has cost you before. The aggregate view counts them, and the count is the point. Compare it to the number you were carrying in your head.

Two wallets, one trader

If you run more than one wallet, a main and a burner, or one for size and one for the fast stuff, each attached wallet gets its own dossier scoped to itself. Free holds one, Pro holds three, Degen holds ten.

The comparison is what you are actually after. If the fast wallet has the better win rate and the worse net PnL, the gap is not in what you picked. It is in what you did after the entry. Same trader, same instincts, two different results. Splitting them apart is the only way to see which half of your behaviour is doing the damage.

The window is part of the number

Every figure on the page is bounded by your plan's history window: 30 days on Free, 90 on Pro, 365 on Degen. Dossier reads the closed swaps sitting on the wallet already, so nothing has to be imported by hand.

That bound matters more than it sounds. A wallet that had a catastrophic March does not appear in a 30-day window. The win rate is not wrong, it is scoped, and it changes the moment the window does. Read the figure with the window attached to it or you are reading a different wallet.

If you want the numbers somewhere else, export to CSV or Obsidian-ready Markdown and do your own arithmetic on them.

What it does not do

  • It does not run on addresses you do not control. No pasting a whale wallet, no copy-trader scanning, no checking your friend's PnL. Research on arbitrary addresses is not live, and neither is the public no-signup roast tool.
  • It does not rank you against other traders. No percentile, no leaderboard, no 0-100 wallet score. Letter grades on trades, plain figures in aggregate.
  • It does not look past your plan's window. On Free, anything older than 30 days is not in the win rate, the net PnL or the top tokens.
  • It does not say whether a wallet is worth copying, or whether to buy or sell anything in it. There are no signals anywhere in Dossier, on this surface or any other.
  • It does not read anything outside Solana mainnet. No other chains, no exchange accounts, no importing trades that happened elsewhere.