Contact your Duck Creek implementation team or partner to obtain API credentials and the base URL for the Duck Creek OnDemand tenant's REST endpoints; Duck Creek exposes 2,600+ REST APIs but documentation is gated behind a partner/customer portal.
Authenticate using OAuth 2.0 client credentials against the tenant's identity provider; store the bearer token as YOUR_DC_TOKEN and refresh before expiry.
Identify the claim lifecycle endpoints (create claim, update claim, close claim) from the Duck Creek Anywhere API catalog provided in your tenant documentation; use standard HTTP verbs GET/POST/PATCH.
Map your internal claim data model to Duck Creek's claim schema, paying attention to required line-of-business-specific fields and enumerated type values defined in your carrier's configuration.
Test end-to-end in the Duck Creek sandbox environment, validating business rules and workflow transitions before promoting to production.
For pre-built third-party connections (MVR, hazard data, document management), use Duck Creek OnDemand productized integrations rather than building custom API calls.
Known gotchas
Duck Creek's API schema is heavily carrier-configured — field names, enumerations, and required fields differ across implementations; never assume a field from one carrier's Duck Creek instance maps to another.
Rate limits and API versioning policies are set at the tenant level and are not publicly documented; negotiate SLAs with your Duck Creek account team before production load testing.
Duck Creek's low-code configuration layer means business rules can silently reject API payloads that pass schema validation; always inspect the response body for business-rule error codes, not just HTTP status codes.
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp