360 Secure SkillHub
360 安全技能中心

Uniswap Monitor Tokenjar

这是Uniswap协议费系统的监控工具,支持一键快照(即时含分析的状态汇总)和实时流(跟踪存款并动态更新)两种模式。核心输出为经多数据复合计算的「下次盈利燃烧预计时间」,还整合TokenJar余额、积累速率、燃烧经济分析及历史记录为结构化仪表盘。适用于用户询问跟踪费积累、下次盈利燃烧时间等场景,输入为相关指令,输出格式化报告或实时更新,可设阈值提醒。

wpank
wpank
热度 47.77
最近更新 2026-05-11 16:09:51
分类:数据依赖:MCP 服务 / 内部调用大模型应用场景:数据处理 / 报告撰写 / 财务

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 monitor-tokenjar
Metadata
name
monitor-tokenjar
description
Monitor the Uniswap TokenJar with a real-time dashboard showing balances, accumulation rates, burn economics, and projected time to next profitable burn. Supports one-shot snapshot and streaming modes. Use when user asks "Watch the TokenJar", "Track fee accumulation", or "When is the next profitable burn?"
model
opus
allowed-tools
["Task(subagent_type:protocol-fee-seeker)","mcp__uniswap__get_tokenjar_balances","mcp__uniswap__get_firepit_state","mcp__uniswap__get_fee_accumulation_rate","mcp__uniswap__subscribe_tokenjar","mcp__uniswap__get_burn_history","mcp__uniswap__get_token_price"]

Monitor TokenJar

Overview

A monitoring dashboard for Uniswap's protocol fee system. The TokenJar accumulates fees from all Uniswap sources (V2, V3, V4, UniswapX, Unichain native fees). This skill provides a comprehensive view of what's in the jar, how fast it's growing, and when the next burn will be profitable -- the single most actionable question for protocol fee seekers.

Two modes: one-shot (instant snapshot with analytics) and streaming (real-time deposit tracking with live updates).

Why this is 10x better than calling tools individually:

  1. Actionable projection: The key output is "estimated time to next profitable burn" -- a compound calculation that requires TokenJar balances, UNI price, gas estimates, and accumulation rates. No single tool provides this. Manually computing it requires calling 4-5 tools and doing the math yourself.
  2. Compound dashboard: Instead of raw JSON from separate tools, you get a single formatted view combining balances, rates, burn economics, and history. The agent cross-references all data sources to produce insights none of them provide alone.
  3. Streaming mode with context: Raw subscribe_tokenjar returns deposit events without context. This skill enriches each deposit with a running total, updated profitability estimate, and alert when the threshold is crossed -- turning raw events into actionable intelligence.
  4. Historical context: The dashboard includes recent burn history and competitive intelligence, so you understand not just the current state but the dynamics of the system.

When to Use

Activate when the user says anything like:

  • "Watch the TokenJar"
  • "Monitor protocol fees"
  • "Track fee accumulation"
  • "When is the next profitable burn?"
  • "Show me TokenJar analytics"
  • "How fast are fees accumulating?"
  • "Alert me when a burn is profitable"
  • "TokenJar dashboard"
  • "What's the accumulation rate?"

Do NOT use when the user wants to execute a burn (use seek-protocol-fees instead) or wants deep historical analysis of burn economics (use analyze-burn-economics instead).

Parameters

Parameter Required Default How to Extract
chain No ethereum Always Ethereum mainnet for TokenJar
streaming No false "watch", "stream", "live", "real-time" implies true
duration No 60 Streaming duration in seconds (1-300). "Watch for 5 minutes" = 300
alert-threshold-usd No -- "Alert me when jar hits $50K" extracts 50000
include-history No true "Skip history" or "just current state" implies false

Workflow

One-Shot Mode (default)

  1. Parallel data collection: Make all MCP calls simultaneously for speed:

    • mcp__uniswap__get_tokenjar_balances -- current jar contents
    • mcp__uniswap__get_firepit_state -- threshold, nonce, readiness
    • mcp__uniswap__get_fee_accumulation_rate -- daily/weekly/monthly rates
    • mcp__uniswap__get_burn_history (if include-history: true) -- recent burns
  2. Compound analysis: Delegate to Task(subagent_type:protocol-fee-seeker) in monitoring mode:

    Produce a TokenJar monitoring dashboard.
    
    Current data:
    - TokenJar balances: {from parallel calls}
    - Firepit state: threshold={threshold}, nonce={nonce}
    - Accumulation rates: {from parallel calls}
    - Recent burn history: {from parallel calls}
    
    Tasks:
    1. Price all TokenJar assets in USD using get_token_price.
    2. Calculate total jar value.
    3. Calculate current burn cost (threshold UNI * UNI price + gas estimate).
    4. Determine current profitability: jar value vs. burn cost.
    5. Using accumulation rates, project when the next burn will be profitable
       (if not already) or when ROI will exceed 10% (if already profitable).
    6. Summarize recent burn history: last burn date, frequency, average profit.
    7. Identify the top fee-generating tokens and any notable trends.
    
    Return a structured dashboard report.
    
  3. Format and present: Display the dashboard with all sections.

Streaming Mode

  1. Initial snapshot: Run the same one-shot workflow above to establish a baseline.

  2. Start streaming: Call mcp__uniswap__subscribe_tokenjar with the user's duration:

    • If alert-threshold-usd is set, include minDepositUsd filter.
    • Default duration: 60 seconds.
  3. Enrich deposits: For each deposit event received:

    • Price the deposited token in USD.
    • Update the running jar total.
    • Recalculate profitability with the new total.
    • If the jar value crosses the burn cost threshold, alert: "Burn is now profitable!"
  4. Final summary: After streaming ends, present an updated dashboard with:

    • Deposits observed during the session.
    • Updated jar total.
    • Updated profitability estimate.

Output Format

One-Shot Dashboard

TokenJar Dashboard (Ethereum)

  ══════════════════════════════════════
  CURRENT BALANCES
  ══════════════════════════════════════
  Token     Balance       USD Value    Share
  WETH      7.20          $18,000      34.6%
  USDC      15,000        $15,000      28.8%
  USDT      8,500         $8,500       16.3%
  WBTC      0.08          $6,400       12.3%
  DAI       4,100         $4,100       7.9%
  ──────────────────────────────────────
  Total                   $52,000      100%

  ══════════════════════════════════════
  ACCUMULATION RATES
  ══════════════════════════════════════
  Daily:    ~$7,400/day
  Weekly:   ~$51,800/week
  Monthly:  ~$222,000/month

  Top Contributors:
    WETH     ~$2,800/day  (37.8%)
    USDC     ~$2,100/day  (28.4%)
    USDT     ~$1,200/day  (16.2%)

  ══════════════════════════════════════
  BURN ECONOMICS
  ══════════════════════════════════════
  Burn Threshold:  4,000 UNI ($28,000)
  Gas Estimate:    ~$45
  Total Burn Cost: $28,045

  Current Jar:     $52,000
  Net Profit:      $23,955
  ROI:             85.4%
  Status:          PROFITABLE

  ══════════════════════════════════════
  RECENT HISTORY
  ══════════════════════════════════════
  Last Burn:       2026-02-03 (7 days ago)
  Burn Frequency:  ~every 5.2 days (avg last 10 burns)
  Avg Profit:      $18,400 per burn
  Nonce:           42

  ══════════════════════════════════════
  PROJECTION
  ══════════════════════════════════════
  Next 10% ROI:    Already exceeded (current: 85.4%)
  Next 100% ROI:   ~0.5 days at current rate
  Competitor Risk:  HIGH — avg burn interval is 5.2 days, currently at 7 days

Streaming Output

TokenJar Live Feed (streaming for 60s)

  Baseline: $52,000 across 5 assets

  [14:00:12] Deposit: 0.15 WETH ($375)  | Running Total: $52,375
  [14:00:28] Deposit: 500 USDC ($500)    | Running Total: $52,875
  [14:00:45] Deposit: 200 USDT ($200)    | Running Total: $53,075

  ──────────────────────────────────────
  Session Summary (60s)
  ──────────────────────────────────────
  Deposits:     3 events, $1,075 total
  Rate:         ~$64,500/hour (this session)
  Updated Total: $53,075
  Profitability: $25,030 net profit (89.3% ROI)
  Status:        PROFITABLE — ready to burn

Alert Output (when threshold crossed)

  ALERT: TokenJar value ($50,125) has crossed your alert threshold ($50,000).
  Current profitability: $22,080 net profit (78.7% ROI).
  Consider running: seek-protocol-fees

Important Notes

  • Read-only skill. This skill never executes transactions. It only reads data and produces analysis. To execute a burn, use seek-protocol-fees.
  • Ethereum mainnet only. The TokenJar and Firepit are mainnet contracts.
  • Accumulation rates are estimates. They are based on historical Transfer events over a lookback window (default ~7 days). Actual rates vary with protocol volume and fee settings.
  • Streaming duration is capped at 300 seconds (5 minutes) by the MCP tool. For longer monitoring, re-run the skill periodically.
  • Competitor intelligence is approximate. Burn frequency is derived from on-chain history, not mempool monitoring. Another searcher could burn at any time.
  • UNI price volatility affects projections. The "time to profitable burn" projection assumes stable UNI price. A UNI price spike could make a currently-profitable burn unprofitable.

Error Handling

Error User-Facing Message Suggested Action
TokenJar empty "TokenJar is empty. No fees have accumulated yet." Wait for protocol activity
No accumulation data "Insufficient data to calculate accumulation rates." Try a larger lookback window
No burn history "No burn history found. This may be a new deployment." Set include-history: false
Streaming timeout "Streaming session ended after {duration}s." Re-run for another session
No deposits during stream "No deposits observed during the {duration}s streaming window." Try a longer duration or check later
Token price unavailable "Could not price {token}. Dashboard values may be incomplete." Token may be exotic or illiquid
RPC connection failed "Cannot connect to Ethereum RPC. Dashboard unavailable." Check RPC configuration