360 Secure SkillHub
360 安全技能中心

Uniswap Optimize Lp

核心功能:依托链上数据与风险分析,为指定代币对生成可落地的LP策略,涵盖版本、费率层级、区间宽度、再平衡方案,搭配收益、滑点及风险评估。典型场景:用户咨询代币对LP策略、版本/费率/区间选择等。输入需提供代币对,可选资金、链、风险偏好;输出为结构化策略报告,含具体方案、预期收益、风险评级及替代选项。

wpank
wpank
热度 55.76
最近更新 2026-05-11 16:09:51
分类:数据依赖:MCP 服务应用场景:财务 / 市场分析 / 推荐

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add optimize-lp
Metadata
name
optimize-lp
description
Get the optimal LP strategy for a token pair — recommends version (V2/V3/V4), fee tier, range width, and rebalance approach based on pair characteristics, historical data, and risk tolerance. Use when the user asks how to LP, what range to use, or which version/fee tier is best.
model
opus
allowed-tools
["Task(subagent_type:lp-strategist)"]

Optimize LP

Overview

Provides a focused, actionable LP strategy recommendation for a specific token pair. Delegates to the lp-strategist agent, which analyzes pair volatility, evaluates version options, selects the optimal fee tier, calculates the best range width, and designs a rebalance strategy — all backed by on-chain data and risk analysis.

This skill answers the question: "How should I LP into X/Y?" with a concrete, implementable answer.

When to Use

Activate when the user asks:

  • "Best LP strategy for ETH/USDC"
  • "How should I LP $10K into ETH/USDC?"
  • "Tight or wide range for WETH/USDC?"
  • "V2 or V3 for this pair?"
  • "What fee tier should I use for UNI/WETH?"
  • "Optimize my LP approach for X/Y"
  • "What range should I set for my ETH/USDC position?"
  • "Should I use concentrated liquidity for this pair?"

Parameters

Parameter Required Default How to Extract
token0 Yes — First token: "ETH", "WETH", "USDC", or 0x address
token1 Yes — Second token
capital No — Amount to LP: "$10K", "5 ETH", "$50,000"
chain No ethereum "ethereum", "base", "arbitrum", etc.
riskTolerance No moderate "conservative", "moderate", "aggressive"

Workflow

  1. Extract parameters from the user's request. Identify the token pair, capital amount (if mentioned), chain, and risk tolerance.

  2. Delegate to lp-strategist: Invoke Task(subagent_type:lp-strategist) with the parameters. The agent performs a 7-step analysis:

    • Classifies the pair (stable-stable, stable-volatile, volatile-volatile)
    • Evaluates V2 vs V3 vs V4 tradeoffs
    • Selects optimal fee tier with data backing
    • Calculates range width targeting >80% time-in-range
    • Designs rebalance strategy with gas cost analysis
    • Gets independent risk assessment from risk-assessor
    • Produces a recommendation with conservative/moderate/optimistic estimates
  3. Present the strategy as a clear, actionable recommendation with specific numbers.

Output Format

LP Strategy for WETH/USDC

  Recommendation: V3, 0.05% fee tier
  Pool: 0x88e6A0c2dDD26FEEb64F039a2c41296FcB3f5640 (Ethereum)

  Range: $1,668 - $2,258 (±15%, medium strategy)
  Expected time-in-range: ~82%

  Expected Returns (annualized):
    Conservative: 8.5% APY (after IL)
    Moderate:     15.2% APY (after IL)
    Optimistic:   24.1% APY (after IL)

  Expected IL:
    Conservative: 2.1%
    Moderate:     5.8%
    Worst case:   12.3%

  Rebalance Strategy:
    Trigger: Price within 10% of range boundary
    Frequency: Every 2-3 weeks (estimated)
    Gas cost: ~$15 per rebalance

  Risk Assessment: MEDIUM (approved by risk-assessor)

  Alternatives Considered:
    V3 0.3%: Lower APY (8%) but less competition
    V2: No range management, ~4% APY — good for passive
    Narrow range (±5%): Higher APY but needs weekly rebalancing

  Ready to add liquidity? Say "Add liquidity to WETH/USDC"

Important Notes

  • This skill provides a strategy recommendation, not execution. To act on it, use the manage-liquidity skill.
  • All APY estimates are based on historical data. Past performance does not guarantee future returns.
  • IL estimates are always shown alongside fee APY — never fee APY alone.
  • For small positions (<$1K), the recommendation accounts for gas costs eating into returns.
  • The lp-strategist internally delegates to pool-researcher for data and risk-assessor for risk evaluation.

Error Handling

Error User-Facing Message Suggested Action
Token not found "Could not find token X on Uniswap." Provide contract address
No pools exist "No pools found for X/Y on this chain." Try different chain or check tokens
Insufficient data "Not enough historical data to produce a reliable strategy." Pool may be too new
Agent unavailable "LP strategist is not available." Check agent configuration