credit_score¶

Returns an AI-driven crypto trust score for a wallet address — a 1–9 rating that assesses borrower reliability by combining on-chain inflow/outflow analysis, fraud probability, and social graph signals.

Designed for DeFi lending protocols that need to differentiate reliable borrowers from high-risk wallets before originating loans — without requiring identity data.

Combines predictive_fraud scoring with social graph analysis in a single call
1–9 scale: 1 = lowest trust, 9 = highest trust
Covers ETH, BNB, POLYGON, TON, BASE, HAQQ
Response latency under 100ms — suitable for real-time credit gating at wallet connect

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

Supported Networks¶

Identifier	Network
`ETH`	Ethereum
`BNB`	BNB Smart Chain
`POLYGON`	Polygon
`TON`	TON
`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	Wallet address to score

Output Schema¶

{
  "message": "Success",
  "creditData": {
    "riskRating": 7,
    "walletAddress": "0x..."
  }
}

`riskRating` Interpretation¶

Score	Trust Level	Recommended Action
`8–9`	High trust	Approve — reliable borrower profile
`6–7`	Above average	Approve with standard terms
`4–5`	Average	Approve with conservative collateral ratio
`2–3`	Below average	Require higher collateral or manual review
`1`	Low trust	Reject or require maximum collateral

Code Examples¶

Node.js¶

const result = await client.callTool({
  name: "credit_score",
  arguments: {
    apiKey: process.env.CHAINAWARE_API_KEY,
    network: "ETH",
    walletAddress: "0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045"
  }
});

// result.creditData.riskRating → 7
// result.creditData.walletAddress → "0xd8dA..."

Python¶

result = await session.call_tool("credit_score", {
    "apiKey": os.environ["CHAINAWARE_API_KEY"],
    "network": "ETH",
    "walletAddress": "0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045"
})

# result["creditData"]["riskRating"] → 7

Example Agent Prompts¶

"What is the credit score for 0xABC123... on Ethereum?"
"Calculate the trust score for this wallet before approving the loan: 0x..."
"Is this borrower reliable? Score their wallet on BNB Chain: 0x..."
"Credit check on 0x... on BASE before originating this undercollateralised loan."

Use Cases¶

DeFi lending — score borrowers before originating undercollateralised or reduced-collateral loans
Undercollateralised lending — use the 1–9 rating to set dynamic collateral ratios per borrower
Borrower tiering — route high-score wallets to preferential rates; flag low-score wallets for manual review
Credit-gated products — restrict access to advanced financial products to wallets above a minimum score
Yield vault access — gate high-yield, high-risk vaults behind minimum credit scores

Combining with Other Tools¶

For a complete borrower picture, combine credit_score with predictive_fraud and predictive_behaviour:

1. credit_score       → trust rating (1–9)
2. predictive_fraud   → fraud probability (0.00–1.00)
3. predictive_behaviour → experience level, DeFi history, intent signals

This three-call pipeline gives you everything needed for a lending decision: whether the wallet is trustworthy, what its fraud risk is, and how experienced and active it is in DeFi.

Error Codes¶

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

predictive_fraud — detailed fraud probability + AML forensics; call alongside credit_score for a full risk picture
predictive_behaviour — full behavioural profile including experience level, DeFi categories, and intent prediction