Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Riskified E-Commerce API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Friskified.com%2Friskified" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Friskified.com%2Friskified" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with Riskified E-Commerce API.
Screen an order for fraud and get an immediate decision with the synchronous decide call
Submit an order for asynchronous fraud review and receive the decision afterwards
Report order updates, cancellations, refunds, and fulfillments so Riskified tracks the order lifecycle
Notify Riskified of a chargeback and report your own final decision on an order
GET STARTED
Report checkout creation and checkout denials, and ask Riskified to advise on authentication
Send customer login, account creation, password reset, and verification events for account-level scoring
Upload historical orders to train fraud models on your past order outcomes
Patterns agents use Riskified E-Commerce API for, with concrete tasks.
★ AI order-review agent with live fraud decisions
An AI agent handling order operations needs a fraud decision before it releases an order to the warehouse. Through Jentic the agent searches for 'screen an order for fraud', loads the Riskified decide operation with its input schema, and posts the order with the HMAC signature and shop-domain header handled at call time. It reads the approve or decline decision from the response and routes the order to fulfillment or manual review.
POST the new order payload to /api/decide, read the returned decision, and hold the order for manual review if it is declined
Keeping fraud review in sync with the order lifecycle
A store that uses asynchronous review creates the order with Riskified, submits it for analysis, and then reports what happens next. Updates, cancellations, partial or full refunds, and fulfillments each have their own call, so Riskified's view of the order matches the store's.
Call POST /api/create for a new order, then POST /api/submit to send it for review, and POST /api/fulfill once the shipment leaves the warehouse
Chargeback reporting
When a payment processor raises a dispute, the merchant notifies Riskified through the chargeback call so the dispute is tied to the original reviewed order. Reporting the merchant's own decision on an order through the decision call closes the loop for orders that were overridden. Together these record the final outcome of each reviewed order.
When a dispute notice arrives, POST the order id and dispute details to /api/chargeback
Account takeover and policy abuse signals
Fraud is not only at checkout. The Account Action calls let a store report customer logins, new accounts, account updates, password resets, verification attempts, logouts, loyalty redemptions, and wishlist changes. Feeding these events gives Riskified the account history it needs to flag a login or redemption that does not fit the customer's pattern.
POST each successful and failed login to /customers/login and each password reset to /customers/reset_password
22 endpoints — jentic publishes the only available openapi specification for the riskified e-commerce api, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
/api/decide
Synchronous fraud decision
/api/create
Create order
/api/submit
Submit order for review
/api/refund
Refund order
/api/fulfill
Fulfill order
/api/chargeback
Report chargeback
/customers/login
Report customer login
/api/decide
Synchronous fraud decision
/api/create
Create order
/api/submit
Submit order for review
/api/refund
Refund order
/api/fulfill
Fulfill order
/api/chargeback
Report chargeback
/customers/login
Report customer login
What agents get from Jentic-routed access to this vendor.
Setup
Wiring Riskified by hand means computing an HMAC-SHA256 signature of every request body with your auth token, adding the shop-domain header, and choosing between separate order-action and account-action hosts. Through Jentic you install once, import Riskified from the API Directory, store the token once, and your agent calls it.
Permission scoping
Riskified carries the order inside the request body, so rules bound which operations your agent may call, not which order it touches. You choose the operations it may call, so lifecycle changes like cancel, refund, or chargeback reporting are not included unless you add them.
Credential isolation
Your Riskified auth token is stored once, encrypted, by your own Jentic One instance and used at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'screen an order for fraud' or 'report a chargeback', and Jentic returns the matching Riskified operation with its input schema so the agent calls the right endpoint without browsing the reference docs.
Alternatives and complements available in the Jentic catalogue.
Specific to using Riskified E-Commerce API through Jentic.
Why is there no official OpenAPI spec for the Riskified E-Commerce API?
Riskified does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call the Riskified E-Commerce API via structured tooling. It is validated against the live API and kept up to date. To run it on your own infrastructure, install Jentic One from its GitHub repo.
What authentication does the Riskified E-Commerce API use?
Per its OpenAPI spec, each request carries an HMAC-SHA256 signature of the request body, computed with your Riskified auth token and sent in the `X-RISKIFIED-HMAC-SHA256` header. Requests also identify your store with the `X-RISKIFIED-SHOP-DOMAIN` header. Through Jentic the auth token is held by your own instance and used at call time, so the agent never sees it.
Can I get a fraud decision on an order immediately with the Riskified E-Commerce API?
Yes. POST /api/decide submits an order for synchronous fraud analysis and returns the decision in the same response. If you prefer asynchronous review, create the order with POST /api/create and send it for analysis with POST /api/submit instead.
What are the rate limits for the Riskified E-Commerce API?
The OpenAPI spec does not specify rate limits. Check the Riskified API reference at https://apiref.riskified.com/ or your Riskified account team for current limits before sending historical order uploads in bulk.
Is there a Riskified MCP server?
You do not need an MCP server to give your agent Riskified. Jentic connects it directly from the API Directory: import it, store your auth token once, and your agent calls the order and account operations it needs on demand without loading another server's tool definitions into its context.
Can I limit what my agent is allowed to do with the Riskified E-Commerce API?
Yes. For an order-review agent, write a rule that allows only POST /api/decide and POST /api/fulfill, so it can screen orders and mark them shipped but cannot cancel, refund, or report chargebacks. Every call it makes is logged.
Does the Riskified E-Commerce API use different hosts for order and account actions?
Yes. The spec lists https://wh.riskified.com for production order actions and https://api.riskified.com for production account actions, with https://sandbox.riskified.com and https://api-sandbox.riskified.com as the matching sandbox hosts. Use the sandbox hosts while testing an integration.
Know of an official OpenAPI document? Contribute it →
For Agents
Submit e-commerce orders to Riskified for fraud screening, get approve or decline decisions, and report refunds, fulfillments, chargebacks, and customer account events.
Use for: Check whether this order should be approved before I ship it, Submit a new order for fraud review, I need to report a refund on an order Riskified already reviewed, Report a chargeback that just came in for an order
Not supported: Does not process payments, issue refunds to customers, or manage store inventory. Use for fraud screening and reporting order and account events to Riskified only.
Jentic publishes the only available OpenAPI specification for the Riskified E-Commerce API, keeping it validated and agent-ready. The Riskified E-Commerce API screens online orders for fraud and returns an approve or decline decision, either synchronously through the decide call or asynchronously after an order is submitted for review. Merchants keep Riskified in step with the order lifecycle by reporting updates, cancellations, refunds, fulfillments, chargebacks, and their own final decisions. A separate group of Account Action calls reports customer logins, account creation, password resets, verification, wishlist changes, and contact events so account takeover and policy abuse can be scored alongside orders.
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
Base layer of spec validity and structural soundness.
Aggregated quality score from linter diagnostics, weighted by severity.
Percentage of `$ref` references that resolve successfully.
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
Structural correctness score based on schema issues using logarithmic dampening.
Clarity, completeness, and ingestion readiness for developers and tooling.
How richly the API is illustrated with examples.
Percentage of examples that conform to their schemas.
Percentage of operations with complete response definitions (success, client error, server error).
Health of API ingestion, bundling, and resolution within Jentic pipelines.
Semantic breadth, depth, and agent comprehension for AI systems.
Coverage of descriptions across API elements.
Coverage of RFC 9457 Problem Details for error responses.
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
Coverage of summaries across operations/tags/info.
Functional utility, complexity comfort, and AI orchestration readiness.
Agent comfort level based on API operational and structural complexity.
Trust, risk posture, and security compliance.
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
Findability, semantic richness, and reasoning readiness.
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>