predictive_rug_pull¶

Forecasts whether a smart contract or liquidity pool is likely to execute a rug pull — before investors deposit capital. Analyses the contract itself, its deployer's cross-chain behavioural history, and the behaviour of associated liquidity providers.

Core insight: bad actors cannot create good contracts. A deployer's behavioural history across 8 chains reveals who they are, regardless of how polished their website or whitepaper looks.

68% detection accuracy on new pools within the first hour of launch
Three-layer scoring: contract + deployer + liquidity
Continuous updates as on-chain activity evolves

MCP Endpoint: https://prediction.mcp.chainaware.ai/sse

Supported Networks¶

Identifier	Network
`ETH`	Ethereum
`BNB`	BNB Smart Chain
`BASE`	Base
`HAQQ`	HAQQ

Input Schema¶

Field	Type	Required	Description
`apiKey`	string	Yes	Your ChainAware API key
`network`	string	Yes	One of the network identifiers above
`walletAddress`	string	Yes	Smart contract address or liquidity pool address to check

Note: pass the contract or pool address here, not a wallet address. To check a wallet address, use predictive_fraud.

Output Schema¶

{
  "message": "string",
  "contractAddress": "string",
  "status": "Fraud | Not Fraud",
  "probabilityFraud": 0.87,
  "lastChecked": "ISO-8601 timestamp",
  "forensic_details": {
    "...": "on-chain metrics used in scoring"
  },
  "createdAt": "ISO-8601 timestamp",
  "updatedAt": "ISO-8601 timestamp"
}

`status` Values¶

Value	Meaning
`Not Fraud`	Low rug pull probability — contract appears legitimate
`Fraud`	High rug pull probability — avoid depositing

`probabilityFraud` Interpretation¶

Range	Risk Level	Recommended Action
`0.00–0.20`	Low	Contract appears safe
`0.21–0.50`	Medium	Proceed with caution; monitor deployer activity
`0.51–0.80`	High	Warn users prominently before they deposit
`0.81–1.00`	Critical	Block listing or deposit entirely

How the Scoring Works¶

The model evaluates three layers simultaneously:

1. Contract layer
Bytecode patterns, proxy structures, admin key permissions, honeypot indicators, mint functions, and ownership concentration. Catches contracts designed to trap funds even when they pass standard audits.

2. Deployer layer
The wallet that deployed the contract is scored for its full cross-chain behavioural history. Serial rug pullers are flagged even when deploying fresh, clean-looking contracts — because their deployment patterns and prior history are visible across chains.

3. Liquidity layer
LP wallet behaviours, lock status, withdrawal velocity patterns, and concentration of liquidity in suspicious or connected addresses.

All three layers feed a single probabilityFraud score, updated continuously as on-chain activity evolves.

Code Examples¶

Node.js¶

const result = await client.callTool({
  name: "predictive_rug_pull",
  arguments: {
    apiKey: process.env.CHAINAWARE_API_KEY,
    network: "BNB",
    walletAddress: "0xContractAddressHere"
  }
});

if (parseFloat(result.probabilityFraud) > 0.5) {
  console.warn("High rug pull risk:", result.probabilityFraud);
  blockDeposit();
} else {
  console.log("Contract appears safe:", result.status);
}

Python¶

result = await session.call_tool("predictive_rug_pull", {
    "apiKey": os.environ["CHAINAWARE_API_KEY"],
    "network": "ETH",
    "walletAddress": "0xContractAddressHere"
})

print(result["status"])           # "Not Fraud"
print(result["probabilityFraud"]) # 0.05

Example Agent Prompts¶

"Will this new DeFi pool rug pull if I stake my assets?"
"Is this smart contract on BNB safe? Address: 0x456..."
"Check the rug pull risk of this ETH liquidity pool."
"Should I invest in this new launchpad project? Contract: 0x789..."
"Scan these 10 contracts before we list them on our DEX."
"Flag any pools on BASE with a rug pull probability above 0.5."

Use Cases¶

Launchpads — vet every project before listing; block high-risk contracts automatically
DEXes — scan new liquidity pools before they appear in your UI
Wallets and portfolio trackers — warn users in real time when they're about to interact with a flagged contract
Individual investors — check any token contract before buying or providing liquidity
DeFi aggregators — filter out high-risk pools from yield routing
Insurance protocols — price smart contract cover based on rug pull probability scores

Error Codes¶

Code	Meaning
`403`	Invalid or missing `apiKey`
`400`	Malformed `network` or `walletAddress`
`500`	Temporary backend failure — retry after a short delay

predictive_fraud — for scoring wallet addresses (not contracts)
token_rank_single — combine with predictive_rug_pull for a full token safety assessment: rug pull risk on the contract + holder community quality