canonical: https://jentic.com/apis/chainaware.ai/chainaware

# ChainAware Enterprise API

ChainAware Enterprise API delivers AI-driven, real-time risk signals for Web3 wallets and contracts, scoring fraud probability for an address, predicting rug-pull risk for a smart contract, segmenting wallets by behaviour, and producing a Web3 credit score. All scoring runs against on-chain activity rather than off-chain KYC, so DeFi protocols and wallet apps can plug in without collecting identity data.

## For AI agents

Score the fraud risk of a wallet, detect rug-pull risk in a contract, segment wallets by behaviour, and produce a Web3 credit score for any blockchain address.

## Scope

Does not perform off-chain KYC, sanctions screening, or transaction broadcast - use for AI risk scoring of wallets and contracts on chain only.

## Capabilities

- Calculate the fraud probability of a wallet address from its on-chain history
- Run a full behavioural audit and intent prediction for a wallet
- Score the rug-pull probability of a token or contract address
- Segment a wallet into a behavioural cohort with a quality score
- Generate an AI-driven Web3 credit score for a wallet

## Use cases

### Wallet Fraud Screening

Score a wallet at signup or before a high-value transaction so a dapp or exchange can step up authentication, hold funds, or block based on risk. POST `/fraud/check` returns a fraud probability for the address, and POST `/fraud/audit` returns a fuller behavioural profile and predicted intent. The scoring uses only on-chain data, so no identity collection is required.

Example prompt: Call POST `/fraud/check` with the wallet address and block the deposit flow if the returned probability exceeds 0.7

### Rug Pull Pre-Listing Check

Screen a token contract before listing it on an exchange, aggregator, or wallet UI. POST `/rug/pull-check` returns a rug-pull probability for the contract based on deployer behaviour, liquidity patterns, and known rug signatures. Used by listing teams who need an automated first-pass filter before a manual review.

Example prompt: Call POST `/rug/pull-check` for each new token contract submitted today and queue any with probability above 0.5 for manual review

### Web3 Credit Scoring for Lending

Power undercollateralised or partially-collateralised DeFi lending by feeding a borrower wallet through a behavioural credit score rather than off-chain bureau data. POST `/users/credit-score` returns an AI-driven trust score, and POST `/segmentation/wallet-segment` classifies the wallet into a behavioural cohort that can drive interest-rate tiers.

Example prompt: Call POST `/users/credit-score` for the borrower wallet, then POST `/segmentation/wallet-segment` to read the cohort, and pass both into the lending decision

### AI Agent DeFi Risk Triage via Jentic

An agent assisting a treasury manager receives a request: 'before we send 50k USDC to this counterparty wallet, tell me if it's risky'. Through Jentic, the agent searches for a wallet-fraud-check operation, loads the input schema, and calls `/fraud/check` with the address using credentials from your Jentic One instance. The agent returns a typed risk verdict without the developer wiring x-api-key handling.

Example prompt: Use Jentic to search 'check fraud risk of a wallet', load the operation backed by POST `/fraud/check`, and execute it with the counterparty address

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/fraud/check` | Calculate fraud probability for a wallet address |
| POST | `/fraud/audit` | Full behavioural profile and intent prediction |
| POST | `/rug/pull-check` | Score rug-pull probability for a contract |
| POST | `/segmentation/wallet-segment` | Classify wallet into a behavioural segment |
| POST | `/users/credit-score` | Generate Web3 credit score for a wallet |

## Key resources

- **Fraud** — Score fraud probability and run behavioural audits on wallet addresses
- **Rug Pull** — Score rug-pull probability for token and contract addresses
- **Segmentation** — Classify wallets into behavioural cohorts with a quality score
- **Credit Score** — Generate an AI-driven Web3 credit score for a wallet

## Why Jentic

- **Setup:** Wiring the ChainAware Enterprise API by hand means handling its x-api-key header and building the request body for each risk-scoring call yourself. Through Jentic you install once, import the ChainAware Enterprise API from the API Directory, store the key once, and your agent calls it.
- **Permission scoping:** ChainAware carries the wallet or contract address in the request body rather than a resource path, so scope your agent to the operations it needs, such as fraud check or rug pull check, and leave the rest out. Every operation you credit it with is one you have added to the allowed set.
- **Credential handling:** Your ChainAware API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'score a wallet for fraud' or 'check rug pull risk', and Jentic returns the matching ChainAware operation with its input schema so the agent posts the right body to the right endpoint without browsing the reference docs.

## Related APIs

- **Alchemy API** — EVM blockchain infrastructure that provides the raw on-chain data ChainAware reasons over
- **Tatum API** — Multi-chain blockchain and wallet API for reading and managing on-chain state
- **AbuseIPDB API** — Reputation scoring for IP addresses rather than wallets

## FAQ

### What authentication does the ChainAware Enterprise API use?

The API uses an API key passed in the x-api-key request header, issued from the ChainAware profile page at chainaware.ai/profile. Through Jentic, the key is held in your Jentic One instance and added to outgoing requests at execution time so it never enters the agent's prompt or context window.

### Can I score a wallet's fraud risk without collecting KYC data?

Yes. POST `/fraud/check` and POST `/fraud/audit` operate against on-chain history alone, so a dapp or exchange can screen wallets at signup or before a transfer without collecting identity documents. The response includes a probability score that can drive automated blocking or step-up flows.

### Does the ChainAware API detect rug pulls before launch?

POST `/rug/pull-check` scores a contract address using deployer behaviour, liquidity patterns, and known rug signatures, so it works on contracts that have just been deployed and on existing tokens. Teams typically use it as a first-pass filter ahead of manual review for listings.

### What are the rate limits for the ChainAware Enterprise API?

Rate limits are tied to the ChainAware enterprise plan attached to your API key rather than published as a single number in the spec. Honour 429 responses with their Retry-After header, especially when batch-scoring large wallet lists.

### How do I get a Web3 credit score for a wallet through Jentic?

Search Jentic for 'get a web3 credit score for a wallet', load the operation backed by POST `/users/credit-score`, and execute it with the wallet address. Jentic injects the x-api-key header so the agent receives a typed score and tier.

### What is the difference between `/fraud/check` and `/fraud/audit`?

POST `/fraud/check` returns a single fraud-probability score suitable for inline decisions. POST `/fraud/audit` returns a fuller behavioural profile and predicted intent for the same wallet, which is more appropriate for case-by-case investigation than for high-throughput screening.

### Can I limit what my agent is allowed to do with the ChainAware Enterprise API?

Yes. Because you run Jentic One yourself, your own rules decide which ChainAware operations the agent may call, so you can grant only the ones it needs, such as POST `/fraud/check` or POST `/rug/pull-check`, and leave the rest out. ChainAware carries the wallet or contract address in the request body rather than a resource path, so scoping is done at the operation level: every operation you add to the allowed set is one the agent can use, and no others. Your ChainAware API key is stored once by your own instance and injected at execution time, so the agent scores addresses without ever handling the credential.