# Rug Pull Detection

## Context

BSC sponsored CertiK to develop a DeFi Project rug pull detection system
https://www.coingecko.com/en/glossary/rug-pulled

## Expected Output

slack notification when we suspect that a project has already rug pulled or has high risk of rug pull

some false positive is acceptable and is better than false negative

## Scope of Projects

All BSC projects

## Information Source

After fact

0. onchain data privileged user transactions **attack pattern**
1. twitter **account doesn't exist**
2. official website **cannot be visited**
3. swap **trading stopped (pancakeswap/uniswap)**
4. coingecko/coin market cap **cannot find project**

### attack analysis

1. mark address to a tag: 0xabc (owner) / attacker
2. mark operations to be function name/params / tx.send value
3. filter transaction by block range: e.g. 100-109 (nice to have)

Before fact

1. reddit **somebody issued warning or very negative sentiment**
2. telegram **very negative sentiment**
3. waronrugs twitter account
4. abnoral APY
5. copycat

### Due Diligence

1. Team (Realworld connection)
2. Workflow

## Examples

- cakelock
- meerkat finance https://bscscan.com/token/0x639f18c72b6a017bdd209c161d1617c9481a1e4d