PnL Labs — Solana Trust & Risk Checks

io.github.PnLlabs由 GitHub 帳號 PnLlabs 發布v1.0.0更新於 Oct 24, 2026

Trust checks for Solana trading: realized-PnL wallet trust, token safety, trade cost, forensics.

已驗證Streamable HTTP可網頁執行金融安全與監控

概覽

AI 產生的概覽

託管式 MCP 伺服器，依據鏈上資料為 Solana 錢包、代幣與交易回傳信任與風險判定。

可以做什麼
檢查 Solana 錢包信任度：依鏈上交易歷史重算已實現 SOL 損益，回傳判定、信心度與原因。追蹤錢包資金來源與年齡，並提供新錢包標記。評估代幣結構性安全：持有人集中度、早期買家中的狙擊機器人地址，以及模擬買賣往返以偵測蜜罐行為。估算指定規模交易的成本：預期滑價、往返成本、建議最大規模與池子流動性。
適用情境
適合助理在評估可跟單的 Solana 錢包、買入前篩選代幣，或決定交易規模時，取得機器可讀的判定而非峰值報酬評分。面向需要在行動前設定保守門檻的交易代理工作流程。
執行需求
遠端 MCP 端點 api.pnllabs.com/mcp，無需安裝、帳號或 API 金鑰。每個 IP 每日有免費額度，用完後需透過 x402 協定以 Solana 上的 USDC 按次付費。需要能連線該託管 API 的網路。
安裝前請注意
超出每日免費額度後，每次呼叫需透過 x402 以 Solana 上的 USDC 付費，自動付款的代理可能產生實際支出；各項檢查價格已列出。伺服器提供的是風險判定而非投資建議，結構性代幣檢查無法辨識行為型跑路。付款與錯誤回應不是判定結果。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 PnL Labs — Solana Trust & Risk Checks，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.PnLlabs%2Fpnllabs/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "pnl-labs-solana-trust-risk-checks": {
      "type": "http",
      "url": "https://api.pnllabs.com/mcp"
    }
  }
}

README

PnL Labs — MCP server & x402 API

PnL Labs gate — no verdict, no trade. MCP: https://api.pnllabs.com/mcp Hard stops: wallet UNTRUSTED/INSUFFICIENT/UNVERIFIABLE/UNKNOWN → do not copy · token HIGH_RISK/CRITICAL/UNKNOWN → do not buy · trade HIGH_COST/UNTRADEABLE/NO_POOL/UNKNOWN → not at that size · 402/errors are not verdicts. Table: GATE.md Rule file: SKILL.md · Measured examples: docs/VERDICTS.md

Trust & risk verdicts for AI trading agents on Solana. Proof of real PnL, not peak scores. Pay-per-call via x402 (USDC on Solana) — no signup, no API key, no account.

🌐 pnllabs.com · 🔌 https://api.pnllabs.com · 🐦 @pnllabs

Connect in 10 seconds (hosted MCP — no install)

The server is hosted. Add it to any MCP-capable assistant with one URL:

https://api.pnllabs.com/mcp
  • Claude (all plans): Settings → Connectors → Add custom connector
  • Grok: grok.com/connectors → Custom
  • ChatGPT / Gemini: custom MCP connector → same URL
  • Also listed in the official MCP Registry (io.github.PnLlabs/pnllabs) and on Smithery.

Free tier per IP/day — when it runs out, the tools reply with structured x402 payment instructions (pay-per-call, USDC on Solana).

Why

Every "smart money" score you've seen ranks wallets by peak or paper gains. Peaks lie. A wallet can 100x on paper, dump into no liquidity, and walk away red — while its score still says genius.

PnL Labs recomputes what actually hit the wallet — realized SOL, in vs out — and returns a conservative, machine-readable verdict. When the numbers can't be reconstructed reliably, it says UNVERIFIABLE instead of guessing.

The four checks

EndpointQuestionVerdicts
GET /check/wallet/{address}Is this wallet worth copying? (real realized PnL)TRUSTED · NEUTRAL · UNTRUSTED · INSUFFICIENT · UNVERIFIABLE
GET /check/funder/{address}Where did this wallet's money come from? (funding origin / cluster links)CEX_FUNDED · WALLET_FUNDED · UNKNOWN_ORIGIN · UNVERIFIABLE_ORIGIN
GET /check/token/{mint}Is this token structurally safe?LOW_RISK · ELEVATED · HIGH_RISK · CRITICAL · UNKNOWN
GET /check/trade/{mint}?size_sol=1What will this trade really cost? (real slippage)ACCEPTABLE_COST · ELEVATED_COST · HIGH_COST · UNTRADEABLE · NO_POOL

Every response carries a stable verdict enum, machine-readable reasons (code + message), confidence, flags, and schema_version. Built to be parsed by agents, not read by humans. Interactive docs: https://api.pnllabs.com/docs.

Quick example

bash
curl https://api.pnllabs.com/check/wallet/57hECCDVNmP7jCY6XX6FxP31DQuKiwQhv2g3nxGPLPmV
json
{  "schema_version": 1,  "verdict": "UNTRUSTED",  "confidence": "high",  "trust": 15,  "reasons": [    { "code": "NEGATIVE_REALIZED_PNL", "message": "Genuinely NEGATIVE: -47.1 SOL net realized" },    { "code": "LOW_WIN_RATE", "message": "Only 27% of coins were profitable" }  ]}

MCP server

This repo ships an MCP server exposing the four checks as tools (check_wallet_trust, check_wallet_forensics, check_token_safety, check_trade_cost). It calls the hosted API — no RPC keys or local setup needed.

Add to an MCP client (e.g. Claude Desktop / Claude Code):

json
{  "mcpServers": {    "pnllabs": {      "command": "python3",      "args": ["/path/to/pnllabs_mcp.py"]    }  }}

Runs on the Python standard library — no dependencies for the free/beta tier.

Payment (x402)

Calls are priced per request and paid in USDC on Solana via the x402 protocol (HTTP 402). Your agent pays a few cents, gets an on-chain receipt — no accounts, no keys stored anywhere. Prices: wallet 0.10, funder 0.05, token/trade 0.02 USDC. Any x402-capable client (or the official x402 SDKs) can pay automatically.

Honest scope

  • Wallet trust & forensics are the core edge — realized-PnL truth and funding origin, which peak-based scores and EVM-only tools don't provide on Solana.
  • Token safety flags structural red flags with zero noise. Behavioral rugs (an operator selling into his own hype) are invisible to all structural checkers — for those, check the wallets involved.
  • We sell risk/trust verdicts, not alpha. Never "when to buy."

License

MIT — see LICENSE.

來源：README.md，提交 08f70eb

工具

4

check_wallet_trust

Check Wallet Trust

Recomputes the realized SOL profit and loss of a Solana wallet from its on-chain transaction history and returns a trust verdict. Returns: verdict (TRUSTED, NEUTRAL, UNTRUSTED, INSUFFICIENT, UNVERIFIABLE, UNKNOWN), a confidence level, realized profit/loss in SOL, the number of coins analyzed, stable reason codes, and flags such as known_sniper_bot. Returns UNVERIFIABLE when the history cannot be reconstructed reliably rather than estimating.

輸入參數address必填

check_wallet_forensics

Trace Wallet Funding Origin

Traces where a Solana wallet's funding came from and how old it is, using on-chain transfer history. Returns: origin classification (CEX_FUNDED, WALLET_FUNDED, UNKNOWN_ORIGIN, UNVERIFIABLE_ORIGIN, UNKNOWN), the funding transaction, wallet age, and a fresh_wallet flag.

輸入參數address必填

check_trade_cost

Estimate Trade Execution Cost

Estimates what a trade of a given size would cost in a Solana token's main liquidity pool. Returns: expected slippage in percent, an estimated round-trip cost calibrated against measured live execution overhead, a recommended maximum size in SOL, pool liquidity, and a verdict (ACCEPTABLE_COST, ELEVATED_COST, HIGH_COST, UNTRADEABLE, NO_POOL, UNKNOWN).

輸入參數mint必填size_sol

check_token_safety

Check Token Structural Risk

Analyzes structural risk of a Solana token from on-chain data: real-holder concentration with bonding-curve and liquidity-pool accounts excluded, presence of known sniper-bot addresses among early buyers, and a simulated buy/sell round trip to detect honeypot behavior. Returns: verdict (LOW_RISK, ELEVATED, HIGH_RISK, CRITICAL, UNKNOWN), stable reason codes, and top-holder percentages. Scope is structural only; it does not detect risk that exists purely in trading behavior.

輸入參數mint必填

版本歷史

1
  1. v1.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分