Whale Index 自动镜像交易策略 | 跟随顶级交易者，智能评分与每日再平衡

whale-index by senpi-ai/senpi-skills

121 周安装量

61 GitHub Stars

GitHub

安装命令

npx skills add https://github.com/senpi-ai/senpi-skills --skill whale-index

自动化区块链金融科技

🇨🇳中文介绍

Whale Index — 自动镜像顶级发现交易者

扫描发现排行榜，为交易者评分，创建镜像策略，并每日进行再平衡。设定您的风险等级和预算 — 代理将处理选择、分配、监控和互换。

技能归属

创建策略时，请在调用中包含 skill_name 和 skill_version。详情请参阅 references/skill-attribution.md。

5 步流程

步骤 1：用户引导

收集：预算、风险承受能力（保守/适中/激进）。

预算	仓位数量
$500-$2k	2
$2k-$5k	3
$5k-$10k	4
$10k+	5

风险映射：

风险等级

广告位招租

在这里展示您的产品或服务

触达数万 AI 开发者，精准高效

联系我们

步骤 2：发现交易者

2a. 拉取候选者： discovery_top_traders(limit=50, timeframe="30d")

2b. 硬性筛选条件：

一致性标签符合风险等级
风险标签符合风险等级
至少 30 天交易记录
不在用户当前投资组合中

2c. 为剩余候选者评分：

score = 0.35 × pnl_rank + 0.25 × win_rate + 0.20 × consistency + 0.10 × hold_time + 0.10 × drawdown

所有组成部分均归一化到 0-100。

2d. 重叠检查： 比较所选交易者之间的活跃持仓。标记持仓重叠度 >50% 的情况。

2e. 分配权重： 基于评分进行加权分配，每个仓位上限为 35%。上限后重新归一化。

步骤 3：展示与确认

向用户展示：交易者地址、排名、标签、胜率、分配金额。执行前等待用户批准。

对于每个仓位：

通过 strategy_create_mirror 创建镜像策略
设置策略级止损（保守 -10%，适中 -15%，激进 -25%）
确认镜像已激活

步骤 5：每日监控（定时任务）

完整的每日检查流程、互换标准和再平衡逻辑，请参见 references/daily-monitoring.md。

互换标准（必须全部满足）：

表现下降： 排名跌出前 50 名或一致性下降或超过 48 小时不活跃或回撤达到历史水平的 2 倍
持续状态： 处于 WATCH 状态连续 2 天以上（通过 watchCount 跟踪）
更好的替代者： 替代者的评分高出 ≥15%
用户策略级止损未触发

关键原则： 2 天的观察期可防止因暂时性下跌而频繁更换。

要退出：关闭所有镜像策略，将资金返还到主钱包。

discovery_top_traders — 交易者排行榜
strategy_create_mirror — 创建镜像策略
strategy_get_clearinghouse_state — 检查持仓
strategy_close_strategy — 清理

镜像策略产生的交易费用与被镜像交易者的活动相同。对于活跃交易者，每日预算费用约为 0.5-1%。

🇺🇸English

Whale Index — Auto-Mirror Top Discovery Traders

Scan the Discovery leaderboard, score traders, create mirror strategies, and rebalance daily. Set your risk level and budget — the agent handles selection, allocation, monitoring, and swaps.

Skill Attribution

When creating a strategy, include skill_name and skill_version in the call. See references/skill-attribution.md for details.

5-Step Flow

Step 1: Onboard the User

Collect: budget, risk tolerance (conservative/moderate/aggressive).

Budget	Slots
$500-$2k	2
$2k-$5k	3
$5k-$10k	4
$10k+	5

Risk mapping:

Risk	Allowed Labels	Max Leverage
Conservative	ELITE only	10x
Moderate	ELITE, RELIABLE	15x
Aggressive	ELITE, RELIABLE, BALANCED	25x

Step 2: Discover Traders

2a. Pull candidates: discovery_top_traders(limit=50, timeframe="30d")

2b. Hard filters:

Consistency label matches risk level
Risk label matches risk level
Min 30d track record
Not already in user's portfolio

2c. Score remaining candidates:

score = 0.35 × pnl_rank + 0.25 × win_rate + 0.20 × consistency + 0.10 × hold_time + 0.10 × drawdown

All components normalized 0-100.

2d. Overlap check: Compare active positions across selected traders. Flag >50% position overlap.

2e. Allocation weighting: Score-weighted allocation with 35% cap per slot. Re-normalize after capping.

Step 3: Present & Confirm

Show the user: trader address, rank, labels, win rate, allocation amount. Wait for approval before executing.

Step 4: Execute

For each slot:

Create mirror strategy via strategy_create_mirror
Set strategy-level stop loss (-10% conservative, -15% moderate, -25% aggressive)
Confirm mirroring is active

Step 5: Daily Monitoring (Cron)

See references/daily-monitoring.md for the complete daily check procedure, swap criteria, and rebalance logic.

Swap criteria (ALL must be true):

Degraded: dropped below rank 50 OR consistency fell OR inactive 48h+ OR drawdown 2× historical
Sustained: WATCH status for 2+ consecutive days (tracked via watchCount)
Better alternative: replacement scores ≥15% higher
User's strategy-level SL not hit

Key principle: The 2-day watch period prevents churn from temporary dips.

Teardown

To exit: close all mirror strategies, return funds to main wallet.

API Dependencies

discovery_top_traders — trader leaderboard
strategy_create_mirror — create mirror strategy
strategy_get_clearinghouse_state — check positions
strategy_close_strategy — teardown

Fee Estimates

Mirror strategies incur the same trading fees as the mirrored trader's activity. Budget ~0.5-1% daily in fees for active traders.

Weekly Installs

113

Repository

senpi-ai/senpi-skills

GitHub Stars

First Seen

Feb 27, 2026

Security Audits

Gen Agent Trust HubPass SocketPass SnykWarn

Installed on

opencode111

gemini-cli111

github-copilot111

codex111

kimi-cli111

amp111

Skills CLI 使用指南：AI Agent 技能包管理器安装与管理教程

43,100 周安装