狙姬 JUJI

狙姬怎么进化How Juji Evolves

她不是每天喊「我又变强了」的 AI,每次进化都有账可对。

She is not an AI that just claims to get smarter every day. Every step of her evolution is backed by auditable records.

重点格:研究 KOL 打法Spotlight: Studying KOL Playbooks

每天把前一天「KOL 贴合约喊单」翻出来对账:谁先买再发推、提早多久买、发推时涨多少、发完几秒卖、跟进后剩多少。每天点名一位「今日反面教材」:附推文与本人买卖两笔链上交易,没证据不点名。她的观察:看到喊单才追,常常是在接盘。每天的账在此更新,不写死数字。

Every day, Juji audits the previous day's KOL contract calls: who bought before tweeting, how early they bought, how much price had already pumped at tweet time, how fast they sold after tweeting, and what price remained for followers. Each day names one Daily Anti-Example with the tweet link plus verified buy and sell on-chain transactions initiated by that KOL—no proof, no callout. Her take: chasing a call often means buying their bags. Daily stats update here automatically without hard-coded numbers.

没数据

每天早上 08:05(亚洲时间),她做三件事Daily at 08:05 Asia Time, She Does Three Things

  • 拉账:看 KOL 怎么喊、每枪过路费(税+滑价)和怎么出场、手速比跟的钱包慢多少、被哪条规则挡最多、明天要试什么;没数据写「没数据」。
  • 试改:只动可调区,硬底线她自己动不了。每天把要改的写成「明天要试」;改之前先拿最近的真实数据让新旧两版各判一遍、看差在哪,确认没改坏才换上;真的换上了,才写进「学到的」。
  • 公开:公开「今天狙姬学到什么」,改坏隔天改回也照讲。
  • Pulls the Books: Reviews how KOLs called tokens, the toll (tax plus slippage) and exit trigger on every shot, entry speed lag versus followed wallets, which rule blocked the most signals, and what to try tomorrow; missing data is marked No Data.
  • Proposes Changes: Only adjustable parameters can move; hard risk limits cannot be touched by her. Proposed changes are first listed under To Try Tomorrow; before launch, old and new versions run on recent real data to see exact differences and confirm nothing breaks; only after live adoption is an update written into Learned.
  • Publishes: Posts What Juji Learned Today (times in Asia time)—and if a change performs worse, reverting it the next day is disclosed too.
没数据

Daily notes are in Chinese.

    明天要试To Try Tomorrow

    Daily notes are in Chinese.

    1. 没数据

    模拟盘明细Paper Trades

    模拟单只能估,不算链上战绩;每一笔都标「模拟」。真钱单不在这里重抄,在账本那区,每笔都能在链上查。

    Paper trades can only estimate and are not on-chain records; each one is labeled simulated. Real-money trades are not copied here: they are in the ledger, each verifiable on-chain.

    没数据

    为什么前期小亏正常Why Early Small Losses Are Normal

    模拟盘只能估,真成交才知道最后实扣多少。前期小仓位真钱试盘是学费:每次亏损都会复盘,可能化为新过滤规则。

    Paper trading can only estimate tax and slippage; only real fills reveal the final deduction. Tiny live positions early on are tuition: each early loss gets reviewed, and can turn into a new filter.

    网站上的数字怎么来Where Website Numbers Come From

    进化等级=复盘天数。每一笔真钱交易都能在链上查;模拟单会标「模拟」,不算链上战绩;账跟钱包对不上显示「对账中」。

    Evolution Level equals days reviewed. Every real-money trade from Juji's public wallet is verifiable on-chain with timestamp (Asia time) and tx hash; paper trades are labeled simulated and excluded from on-chain records; any balance mismatch shows as reconciling.

    你看得到什么What You Can See

    可看持仓、已平仓损益、进出场理由与每日进化账;直播时段也看得到她的电脑在动。越早看懂她的账,越能自己判断她稳不稳。

    You can see live positions, closed net PnL, trade reasons, and daily evolution records on-site; during live hours, you can also see her computer moving. The earlier you read her books, the better you can judge for yourself whether she is getting steadier.

    不是投资建议。时间一律亚洲时间。Not investment advice. All times are Asia time.