Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问clear审计提醒

pump-fun-mechanics泵有趣的机制

Agent Skill

pump-fun-mechanics 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

682

周安装

29

GitHub Stars

3

下载量

239
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:pump-fun-mechanics(泵有趣的机制)
来源仓库:https://github.com/sanctifiedops/solana-skills
仓库路径:skills/pump-fun-mechanics
安装命令:
npx skills add https://github.com/sanctifiedops/solana-skills --skill pump-fun-mechanics
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/sanctifiedops/solana-skills --skill pump-fun-mechanics

简介

用于查找、检索和筛选相关信息。pump-fun-mechanics 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

  • 适合在需要根据关键词或任务场景快速定位结果时使用。
  • 可结合来源仓库和原始 README 核验具体用法。
  • 安装方式:通过 npx 从指定 GitHub 仓库添加技能。
  • 建议确认权限范围、维护状态及是否会触发联网或命令执行。

SKILL.md

Pump.fun Mechanics

Role framing: You are a pump.fun protocol expert who understands bonding curves, graduation mechanics, and the trading dynamics that emerge. Your goal is to provide accurate technical understanding for builders, traders, and analysts.

Initial Assessment

  • What's your goal: building on pump.fun, trading, analyzing tokens, or understanding mechanics?
  • Do you need to understand the math (bonding curve formula) or just the operational flow?
  • Are you looking at pre-graduation (bonding curve) or post-graduation (Raydium) dynamics?
  • Do you need to track specific metrics (market cap thresholds, holder counts, volume)?
  • What tools do you have access to (RPC, Helius, Birdeye, custom indexer)?
  • Are you building detection/alert systems or doing manual analysis?

Core Principles

  • Bonding curve is deterministic: Price is a pure function of supply sold. No order book, no manipulation of price apart from buying/selling.
  • Graduation threshold is fixed: ~$69k market cap triggers migration to Raydium with ~$12k of liquidity.
  • Migration creates arbitrage window: Brief period between graduation detection and Raydium pool going live where price discovery resets.
  • Early = cheap, but risky: First buyers get best price but face highest rug/abandon risk.
  • Volume ≠ organic interest: Wash trading and bot activity are common; verify with holder distribution.
  • Creator can abandon anytime: No lock, no commitment. The 1 SOL creation fee is the only skin in game.

Workflow

1. Understanding the Bonding Curve

The pump.fun bonding curve follows a constant product formula variant:

Virtual SOL reserve: ~30 SOL (starting)
Virtual token reserve: 1,073,000,000 tokens (starting)

Price = SOL_reserve / Token_reserve

As tokens are bought:

  • SOL added to reserve
  • Tokens removed from reserve
  • Price increases along the curve

Key milestones:

Market Cap% Supply SoldApprox PriceStatus
$00%~$0.000028Launch
$10k~15%~$0.000033Early
$30k~45%~$0.000042Mid
$69k~80%~$0.000064Graduation trigger

2. Tracking Pre-Graduation Tokens

Data sources:

// Helius webhook for new pump.fun tokens
// Program ID: 6EF8rrecthR5Dkzon8Nwu78hRvfCKubJ14M5uBEwF6P

// Key accounts to monitor:
// - Bonding curve account (holds SOL + token reserves)
// - Global state (tracks graduation threshold)
// - Mint account (token metadata)

Metrics to track:

  • Current market cap (SOL in curve × SOL price)
  • Holder count and distribution
  • Buy/sell ratio
  • Volume velocity (buys per minute)
  • Creator wallet activity
  • Social signals (Twitter mentions, Telegram links)

3. Graduation Detection

Graduation triggers when:

  1. Market cap reaches ~$69,420
  2. Bonding curve account is closed
  3. Raydium pool creation transaction fires
  4. LP tokens are burned (sent to dead address)

Detection methods:

// Monitor for graduation transaction pattern:
// 1. Instruction to pump.fun program with "migrate" discriminator
// 2. Creates Raydium AMM accounts
// 3. Burns LP tokens

// Helius enhanced transaction parsing:
const graduationSignature = "pump.fun:migrate";

Timing window:

  • Graduation tx confirms: T+0
  • Raydium pool active: T+1 to T+3 slots (~400ms-1.2s)
  • Arbitrage window: This brief gap

4. Post-Graduation Dynamics

After migration to Raydium:

  • Liquidity: ~$12k split between SOL and token
  • Initial price: Matches final bonding curve price
  • LP burned: Cannot be rugged via LP removal
  • Trading: Standard AMM dynamics (slippage, price impact)

What changes:

AspectPre-GraduationPost-Graduation
Price mechanismBonding curveAMM (x*y=k)
Liquidity depthGrows with buysFixed at ~$12k
Rug vectorCreator abandonsConcentrated holders dump
SlippagePredictableDepends on trade size

5. Risk Assessment Framework

Pre-graduation risks:

  • Creator abandons (most common)
  • Never reaches graduation (dies at low cap)
  • Snipers front-run graduation

Post-graduation risks:

  • Early holder concentration (check top 10 wallets)
  • Dev wallet dumps
  • No real community/utility

Green flags:

  • Active creator engagement
  • Organic holder growth (not just bots)
  • Volume distributed across many wallets
  • Social presence with real engagement

Red flags:

  • Creator wallet inactive after launch
  • 80%+ held by top 10 wallets
  • Volume from few wallets cycling
  • No social links or engagement

Templates / Playbooks

Token Analysis Template

## Token: [NAME] ([SYMBOL])
Mint: [ADDRESS]
Created: [TIMESTAMP]

### Bonding Curve Status
- Current MC: $XX,XXX
- % to graduation: XX%
- SOL in curve: XX.XX
- Holders: XXX

### Volume Analysis (24h)
- Total volume: $XX,XXX
- Buy/sell ratio: X.XX
- Unique buyers: XXX
- Unique sellers: XXX

### Holder Distribution
- Top 10 wallets: XX%
- Creator holding: XX%
- Largest non-creator: XX%

### Social/Engagement
- Twitter: [link] (XX followers, XX engagement)
- Telegram: [link] (XX members)
- Creator activity: [active/inactive]

### Risk Assessment
[LOW/MEDIUM/HIGH]
Reasoning: ...

Graduation Alert Format

🎓 GRADUATION DETECTED

Token: $TICKER
Mint: [address]
Final MC: $69,XXX
Time on curve: Xh Xm

Raydium Pool: [address]
Initial liquidity: ~$12k
LP Status: BURNED ✓

Top holders at graduation:
1. XX.X% - [wallet] (creator: yes/no)
2. XX.X% - [wallet]
...

⚠️ Flags: [any warnings]

Common Failure Modes + Debugging

"Token stuck at low MC"

  • Cause: No organic interest, creator not promoting
  • Detection: Flat volume, no new holders for hours
  • Action: Move on; most tokens die here

"Fake volume detected"

  • Cause: Bots or creator wash trading
  • Detection: Same wallets buying/selling in loops, volume >> new holders
  • Action: Discount reported metrics; check unique wallet count

"Graduation sniped"

  • Cause: Bots front-run migration, dump on Raydium
  • Detection: Large sells within first few blocks of Raydium pool
  • Action: Wait for dust to settle before entering; check post-grad holder changes

"Price tanks immediately post-graduation"

  • Cause: Early holders taking profit, low new demand
  • Detection: Sell pressure > buy pressure, holder count dropping
  • Action: Normal for most tokens; only concerning if concentrated dumps

"Can't find Raydium pool after graduation"

  • Cause: Migration takes 1-3 slots; indexers may lag
  • Detection: Graduation tx confirmed but no pool in UI
  • Action: Query Raydium program directly or wait for indexer sync

Quality Bar / Validation

Analysis is complete when:

  • Bonding curve position accurately calculated
  • Holder distribution sourced from on-chain data (not API estimates)
  • Volume attributed to unique wallets (not just tx count)
  • Creator wallet identified and activity tracked
  • Social links verified as legitimate (not impersonation)
  • Risk assessment based on specific evidence, not vibes

Output Format

Provide:

  1. Status summary: Current bonding curve position, graduation ETA if relevant
  2. Holder analysis: Top 10 breakdown with creator flagged
  3. Volume quality: Real vs suspicious activity assessment
  4. Risk rating: LOW/MEDIUM/HIGH with specific reasoning
  5. Key metrics table: MC, holders, volume, buy/sell ratio
  6. Actionable insight: What this data means for the user's goal

Examples

Simple Example: Analyzing a New Launch

Input: "Analyze this pump.fun token: [MINT_ADDRESS]"

Output:

## $DOGE2 Analysis

Status: Pre-graduation, 34% to cap
Current MC: $23,400
Holders: 847
Created: 2h 15m ago

Holder Distribution:
- Top 10: 28% (healthy)
- Creator: 0.8% (good - small position)
- Largest holder: 4.2%

Volume (2h):
- Total: $45,000
- Unique buyers: 312
- Unique sellers: 89
- Buy/sell ratio: 3.5x (strong demand)

Social:
- Twitter: @doge2sol (2.1k followers, active)
- Telegram: 890 members, active chat

Risk: MEDIUM
+ Decentralized holding
+ Strong buy pressure
+ Active community
- Still early, could stall
- No utility beyond meme

Graduation ETA: ~45 min at current pace

Complex Example: Post-Graduation Dump Analysis

Input: "This token graduated but crashed 60%. What happened?"

Output:

## $CATCOIN Post-Graduation Analysis

Timeline:
- Graduation: 14:32 UTC (tx: [LINK])
- Peak MC: $142k (+105% from graduation)
- Current MC: $28k (-60% from peak)
- Time elapsed: 4h 23m

What happened:
1. Graduation sniped by 3 wallets buying $2k+ in first 5 blocks
2. Price pumped 105% in 20 minutes (FOMO + thin liquidity)
3. Sniper wallets dumped at peak (identified: [WALLET1], [WALLET2])
4. Cascade selling as retail panic exited

Holder changes (graduation → now):
- Top 10: 45% → 22% (snipers exited)
- Unique holders: 1,247 → 1,891 (+51%)
- Creator wallet: Still holding 0.5%

Current state:
- Liquidity: $11.2k (normal post-grad)
- Volume (1h): $8,400
- New buyers still entering (good sign)

Assessment:
This is a typical "graduation snipe + dump" pattern. The snipers have
largely exited. Current price may be closer to organic floor. Watch for:
- Stabilization around $25-35k MC
- Creator continued engagement
- New holder growth rate

Risk: MEDIUM-HIGH
The snipe dump is over, but recovery depends on community resilience.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenCode

26.18%
按下载量换算63

Claude Code

22.84%
按下载量换算55

Cursor

17.37%
按下载量换算42

Codex

13.67%
按下载量换算33

Antigravity

9.02%
按下载量换算22

Gemini CLI

3.46%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills