Back to evaluations
Draft evaluation

LangChain Governance Guard: Auditable Agentic Insurance Workflows — evaluation

Evaluates compliance rate and audit trail completeness across stateful agent steps.

Evaluation type
task based
Challenge
LangChain Governance Guard: Auditable Agentic Insurance Workflows
Difficulty
Intermediate
Rigor
Not declared

The author has not specified a rigor level.

Open linked challenge

Evaluation overview

How the linked challenge is judged: tasks, benchmarks, and criteria count.

Tasks
1
Benchmarks
0
Criteria
0

Task templates

Inputs and expected outputs.

Task 1

governed_claim_processing

Processes claim and verifies whether required compliance escalation gates were triggered correctly.

Input format

JSON containing claim_amount_usd, policy_limits, loss_description

Output format

JSON containing approved (boolean), escalation_triggered (boolean), audit_trail_steps (array)