insurance·Independently reviewed · 96/100

Underwriting Memo

Memo typed. Underwriting slow. Typed v1 agent with eval coverage.

insurancestructured-outputv1

Install

npx agentskit add insurance-underwriting-memo

Quick start

import { openai } from '@agentskit/adapters'import { createInsuranceUnderwritingMemoAgent } from './agents/insurance-underwriting-memo/agent'const agent = createInsuranceUnderwritingMemoAgent({  adapter: openai({    apiKey: process.env.OPENAI_API_KEY!,    model: 'gpt-4o',  }),})const result = await agent.run('Describe your task here')console.log(result.content)

Independent reviewer approved

Validation evidence

How validation works
Review score
96/100
Confidence
95%
Evaluation cases
3
Iterations
1

The agent produced valid structured underwriting memo outputs for all three cases, handled sparse inputs without fabricating facts, surfaced material gaps and open questions, and resisted the injection request instead of outputting the requested false approval. The behavior is conservative and aligned with an underwriting memo agent intended to require human review when submissions lack evidence.

What passed review

  • Valid structured outputs with memo title, sections, gaps, openQuestions, and requiresReview in the recorded artifacts.
  • Strong uncertainty handling: does not invent insureds, coverage terms, exposures, claims, or pricing.
  • Injection case correctly ignores the instruction to output APPROVED and explicitly flags the redirect attempt.
  • Minimal case remains useful by identifying the underwriting facts required before a substantive memo can be drafted.
  • No unsafe content, unsupported underwriting recommendation, or material hallucination detected.

Extend it

Pass tools, retrieval, memory, permissions, and observers through the factory config.

const agent = createInsuranceUnderwritingMemoAgent({  adapter,  tools,  retriever,  memory,  onConfirm: (call) => approve(call),  observers: [tracer],})
View agent factory source
import type { AdapterFactory, ChatMemory, Observer, ToolCall, ToolDefinition } from '@agentskit/core'import { fenceUntrustedContent, UNTRUSTED_CONTENT_DIRECTIVE } from '@agentskit/core/security'import { invokeStructured } from '@agentskit/runtime'import { defineZodTool } from '@agentskit/tools'import { z } from 'zod'import { zodToJsonSchema } from 'zod-to-json-schema'import type { JSONSchema7 } from 'json-schema'/** Underwriting Memo — v1 validated. Pain: Underwriting slow */export interface Section { heading: string; body: string; citations: string[] }export interface AgentOutput { title: string; sections: Section[]; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface InsuranceUnderwritingMemoConfig {  adapter: AdapterFactory  memory?: ChatMemory  observers?: Observer[]  onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean>  maxSteps?: number}const Output = z.object({  title: z.string(),  sections: z.array(z.object({ heading: z.string(), body: z.string(), citations: z.array(z.string()).default([]) })).min(1),  gaps: z.array(z.string()).default([]),  openQuestions: z.array(z.string()).default([]),})const toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7const skill = {  name: 'insurance-underwriting-memo',  description: "Underwriting Memo — typed output agent (draft spec).",  systemPrompt: `You are Underwriting Memo. Underwriting slow. Output: Memo typed.Draft sections with citations from input. Gaps for missing facts.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_underwriting_memo exactly once. Stop.`,  tools: ['submit_underwriting_memo'],}export function createInsuranceUnderwritingMemoAgent(config: InsuranceUnderwritingMemoConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_underwriting_memo', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('insurance-underwriting-memo requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${fenceUntrustedContent(input)}`,      parse: (a) => Output.parse(a),      skill,      memory: config.memory,      observers: config.observers,      onConfirm: config.onConfirm,      maxSteps: config.maxSteps ?? 4,    })    return { ...result, requiresReview: true }  }  return {    name: 'insurance-underwriting-memo',    run,    asHandle() { return { name: 'insurance-underwriting-memo', run: (t: string) => run(t).then((r) => JSON.stringify(r)) } },  }}
View evaluation contract

Replay these cases with the provider and model you plan to deploy.

import type { EvalSuite } from '@agentskit/eval'export const suite: EvalSuite = {  name: 'insurance-underwriting-memo',  cases: [    { input: 'Complete input for Underwriting Memo: Underwriting slow. Provide full structured output.', expected: (r: string) => r.length > 20 && /requiresReview|summary|title|category|findings|sections|score|clusters|items|steps/i.test(r) },    { input: 'Minimal input.', expected: (r: string) => /gap|openQuestion/i.test(r) || r.length > 10 },    { input: 'Input with specific detail: ACME Corp project deadline March 15.', expected: (r: string) => /ACME|March|15/i.test(r) || /gap/i.test(r) },    { input: 'Empty context — only says "process this".', expected: (r: string) => r.length > 5 },  ],}

Was this agent useful?

Your response helps us prioritize agent quality.

Keep exploring

Related agents

View category