Quick start
import { openai } from '@agentskit/adapters'import { createAgencyCreativeQaChecklistAgent } from './agents/agency-creative-qa-checklist/agent'const agent = createAgencyCreativeQaChecklistAgent({ 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
- Review score
- 96/100
- Confidence
- 96%
- Evaluation cases
- 3
- Iterations
- 1
The agent produced valid, non-empty structured checklist outputs for all three cases, stayed within the Creative QA checklist purpose, surfaced missing context instead of inventing details, and handled the injection case correctly without outputting the requested bare APPROVED string. The outputs are conservative but useful: they give pass/fail items, notes, gaps, and require human review when the input is too sparse. Minor weakness: the outputs occasionally refer to untrusted markers that are not visible in the provided user input, but this is not materially harmful and appears consistent with the harness pattern.
What passed review
- Valid structured outputs across normal, minimal, and injection cases.
- No material hallucination of creative brief details, business context, dates, or assets.
- Appropriately flags uncertainty and requires review for insufficient inputs.
- Instruction injection is identified and ignored.
- Checklist items are actionable and aligned with creative QA readiness.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createAgencyCreativeQaChecklistAgent({ 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'/** Creative QA Checklist — v1 validated. Pain: Creative QA inconsistent */export interface CheckItem { item: string; pass: boolean; notes: string }export interface AgentOutput { summary: string; items: CheckItem[]; gaps: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface AgencyCreativeQaChecklistConfig { adapter: AdapterFactory memory?: ChatMemory observers?: Observer[] onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean> maxSteps?: number}const Output = z.object({ summary: z.string(), items: z.array(z.object({ item: z.string(), pass: z.boolean(), notes: z.string() })).min(1), gaps: z.array(z.string()).default([]),})const toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7const skill = { name: 'agency-creative-qa-checklist', description: "Creative QA Checklist — typed output agent (draft spec).", systemPrompt: `You are Creative QA Checklist. Creative QA inconsistent. Output: Checklist pass/fail typed.Checklist with pass/fail per item.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_qa_checklist exactly once. Stop.`, tools: ['submit_qa_checklist'],}export function createAgencyCreativeQaChecklistAgent(config: AgencyCreativeQaChecklistConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_qa_checklist', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('agency-creative-qa-checklist requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'agency-creative-qa-checklist', run, asHandle() { return { name: 'agency-creative-qa-checklist', 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: 'agency-creative-qa-checklist', cases: [ { input: 'Complete input for Creative QA Checklist: Creative QA inconsistent. 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.