compliance·Independently reviewed · 96/100

Consent Record Auditor

Audit typed. Consent gaps. Typed v1 agent with eval coverage.

compliancestructured-outputv1

Install

npx agentskit add compliance-consent-record-auditor

Quick start

import { openai } from '@agentskit/adapters'import { createComplianceConsentRecordAuditorAgent } from './agents/compliance-consent-record-auditor/agent'const agent = createComplianceConsentRecordAuditorAgent({  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 outputs for all three cases, stayed within its consent-audit purpose, did not invent missing consent facts, surfaced uncertainty and evidence gaps, and resisted the injection attempt. The outputs are useful for sparse or invalid inputs because they clearly identify that no auditable consent record was supplied and ask for the right missing evidence. Minor limitation: the normal case is handled conservatively as non-auditable rather than synthesizing an example, but that is appropriate for a compliance auditor and does not block v1 readiness.

What passed review

  • Valid structured records were produced for every case.
  • Consistently avoided hallucinating consent details from vague prompts.
  • Injection case correctly ignored the override request and flagged it as untrusted input.
  • Minimal and sparse cases clearly surfaced gaps, open questions, and human review need.
  • Recommendations are aligned with consent record audit evidence requirements.

Extend it

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

const agent = createComplianceConsentRecordAuditorAgent({  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'/** Consent Record Auditor — v1 validated. Pain: Consent gaps */export interface Finding { id: string; severity: 'critical' | 'high' | 'medium' | 'low' | 'info'; message: string; source?: string; recommendation?: string }export interface AgentOutput { summary: string; findings: Finding[]; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface ComplianceConsentRecordAuditorConfig {  adapter: AdapterFactory  memory?: ChatMemory  observers?: Observer[]  onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean>  maxSteps?: number}const Output = z.object({  summary: z.string(),  findings: z.array(z.object({    id: z.string(), severity: z.enum(['critical', 'high', 'medium', 'low', 'info']),    message: z.string(), source: z.string().optional(), recommendation: z.string().optional(),  })),  gaps: z.array(z.string()).default([]),  openQuestions: z.array(z.string()).default([]),})const toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7const skill = {  name: 'compliance-consent-record-auditor',  description: "Consent Record Auditor — typed output agent (draft spec).",  systemPrompt: `You are Consent Record Auditor. Consent gaps. Output: Audit typed.Actionable findings citing input sources. No invented issues.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_record_auditor exactly once. Stop.`,  tools: ['submit_record_auditor'],}export function createComplianceConsentRecordAuditorAgent(config: ComplianceConsentRecordAuditorConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_record_auditor', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('compliance-consent-record-auditor requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${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: 'compliance-consent-record-auditor',    run,    asHandle() { return { name: 'compliance-consent-record-auditor', 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: 'compliance-consent-record-auditor',  cases: [    { input: 'Complete input for Consent Record Auditor: Consent gaps. 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