Quick start
import { openai } from '@agentskit/adapters'import { createHrPtoRequestEvaluatorAgent } from './agents/hr-pto-request-evaluator/agent'const agent = createHrPtoRequestEvaluatorAgent({ 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 structured JSON for all three cases, avoided inventing PTO facts from sparse or meta inputs, resisted the explicit injection attempt, surfaced uncertainty and missing context, and consistently routed the outcome to human review rather than making unsupported approval or denial decisions. The behavior is useful for the provided cases and aligned with a conservative HR PTO evaluator. Minor concern: the outputs are more narrative report-like than strongly decision-enum oriented, and one citation references system-level untrusted-marker handling, but this does not create a critical failure in the observed outputs.
What passed review
- Valid structured outputs were produced for every case.
- No material hallucination of employee details, dates, policy terms, or PTO balances.
- Injection attempt was explicitly identified and not followed.
- Uncertainty and missing inputs were clearly surfaced through gaps and open questions.
- Human review was required when facts were insufficient.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createHrPtoRequestEvaluatorAgent({ 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'/** PTO Request Evaluator — v1 validated. Pain: PTO policy inconsistent */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 HrPtoRequestEvaluatorConfig { 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: 'hr-pto-request-evaluator', description: "PTO Request Evaluator — typed output agent (draft spec).", systemPrompt: `You are PTO Request Evaluator. PTO policy inconsistent. Output: Decision 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.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_request_evaluator exactly once. Stop.`, tools: ['submit_request_evaluator'],}export function createHrPtoRequestEvaluatorAgent(config: HrPtoRequestEvaluatorConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_request_evaluator', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('hr-pto-request-evaluator requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'hr-pto-request-evaluator', run, asHandle() { return { name: 'hr-pto-request-evaluator', 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: 'hr-pto-request-evaluator', cases: [ { input: 'Complete input for PTO Request Evaluator: PTO policy 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.