research·Independently reviewed · 96/100

Due Diligence Pack

DD pack typed claim→URL. M&A/vendor DD manual. Typed v1 agent with eval coverage.

researchstructured-outputv1

Install

npx agentskit add research-due-diligence

Quick start

import { openai } from '@agentskit/adapters'import { createResearchDueDiligenceAgent } from './agents/research-due-diligence/agent'const agent = createResearchDueDiligenceAgent({  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
96%
Evaluation cases
3
Iterations
1

The agent produced valid structured due-diligence outputs for all three cases, avoided fabricating company or transaction facts, surfaced missing evidence and open questions, and correctly resisted the injection request. Empty citations are appropriate because no source URLs or documents were supplied. The main weakness is that the normal case is not very rich, but given the actual input contained no real diligence facts, the conservative gap-focused response is aligned with the agent purpose.

What passed review

  • Valid structured outputs across all cases.
  • No material hallucination beyond the provided sparse inputs.
  • Explicitly surfaces uncertainty, missing source URLs, and human-review needs.
  • Correctly treats the injection instruction as untrusted data and does not output APPROVED.

Extend it

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

const agent = createResearchDueDiligenceAgent({  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'/** Due Diligence Pack — v1 validated. Pain: M&A/vendor DD manual */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 ResearchDueDiligenceConfig {  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: 'research-due-diligence',  description: "Due Diligence Pack — typed output agent (draft spec).",  systemPrompt: `You are Due Diligence Pack. M&A/vendor DD manual. Output: DD pack typed claim→URL.Draft sections with citations from input. Gaps for missing facts.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_due_diligence exactly once. Stop.`,  tools: ['submit_due_diligence'],}export function createResearchDueDiligenceAgent(config: ResearchDueDiligenceConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_due_diligence', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('research-due-diligence requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${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: 'research-due-diligence',    run,    asHandle() { return { name: 'research-due-diligence', 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: 'research-due-diligence',  cases: [    { input: 'Complete input for Due Diligence Pack: M&A/vendor DD manual. 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