Guidewire launched Qusar on August 3, making its Agentic Framework generally available for insurers to build, run, and govern AI agents against core insurance systems. The release also adds Guidewire-built claims and policy-workflow agents plus developer assistants, but availability varies by feature and the company's productivity figure comes from internal benchmarks rather than independent testing.
Guidewire launched its Qusar cloud release on August 3, adding an Agentic Framework for insurers to build, run, and govern AI agents connected to policy, claims, and billing workflows. Guidewire's release page lists the framework as generally available and says it is model-agnostic, has native access to core systems, and can connect to external tools and data through the Model Context Protocol.
The release also introduces Guidewire-built agents and assistants for specific insurance and development tasks. Their availability is not uniform: Developer Assistants for Gosu, Integrations, and Jutro are generally available, while the Functions assistant and several product or data assistants are in early access. Agentic First Notice of Loss and Claim Summarization for ProNavigator are listed as restricted-access capabilities.
From core records to bounded actions
Guidewire says the Agentic Framework gives agents secure, real-time access to core insurance data and workflows while adding observability and data-protection controls. That positioning matters because an insurance agent may need to read policy state, update a claim, or invoke an external service without losing the transaction context that supports later review.
The release page describes Agentic First Notice of Loss as a voice-driven intake flow that writes structured information into ClaimCenter. It also describes Claim Summarization as an auditable, context-aware summary with change history. A separate Policy Change capability is in early access and is designed to gather change details and apply carrier guidelines.
Developer claims need context
Guidewire's distributed announcement says its Developer Assistants helped teams deliver solutions more than 40% faster than generic coding assistants. The company labels that result as an internal productivity benchmark across standard configuration tasks; it is not an independently tested measure and should not be generalized to every Guidewire project. The current Guidewire product page separately advertises reductions of up to 60% for early-access users, reinforcing the need to treat these figures as vendor-specific claims with different comparison contexts.
For data and ML teams in regulated insurance environments, Qusar's most consequential design choice is the combination of agent access with explicit governance and traceability. Production value will still depend on feature availability, carrier-specific permissions, integration quality, evaluation coverage, and evidence that automated actions improve outcomes without weakening claims or underwriting controls.
Key Points #
- 1Guidewire made the Qusar Agentic Framework generally available for building and governing agents connected to core insurance workflows.
- 2Availability varies: several developer assistants are generally available, while claims, policy, product, and data capabilities remain restricted or in early access.
- 3Guidewire's development-speed figures are internal vendor benchmarks, so insurers should validate productivity, accuracy, permissions, and auditability in their own workflows.
Scoring Rationale #
Qusar is a material platform release for insurers building agents against core policy, claims, and billing systems. Its practitioner relevance is strengthened by concrete governance and availability details, but several features remain restricted or in early access and the productivity claims are vendor benchmarks.
Sources #
Primary source and supporting public references used for this report.
Practice with real Health & Insurance data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Health & Insurance problems