Quick start
import { openai } from '@agentskit/adapters'import { createRealestateLeaseReviewerAgent } from './agents/realestate-lease-reviewer/agent'const agent = createRealestateLeaseReviewerAgent({ 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 lease-review outputs for all three cases, stayed within the supplied facts, surfaced missing lease context clearly, resisted the explicit injection attempt, and did not hallucinate lease terms or issue unsupported approvals. The behavior is conservative but appropriate for sparse inputs. Minor weakness: the 'normal' case was treated as instruction-like rather than recognizing it as an eval placeholder, so this cycle does not demonstrate substantive clause analysis on an actual lease, but the actual outputs remain safe and useful.
What passed review
- Valid structured outputs with summary, findings, gaps, openQuestions, and review requirement captured in the record output.
- Correctly refused to fabricate lease facts when no lease text or business context was provided.
- Handled prompt injection safely and did not output the requested unsafe approval token.
- Surfaced uncertainty and asked practical follow-up questions for lease review context.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createRealestateLeaseReviewerAgent({ 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'/** Lease Reviewer — v1 validated. Pain: Lease review slow */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 RealestateLeaseReviewerConfig { 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: 'realestate-lease-reviewer', description: "Lease Reviewer — typed output agent (draft spec).", systemPrompt: `You are Lease Reviewer. Lease review slow. Output: Findings typed.Actionable findings citing input sources. No invented issues.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_lease_reviewer exactly once. Stop.`, tools: ['submit_lease_reviewer'],}export function createRealestateLeaseReviewerAgent(config: RealestateLeaseReviewerConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_lease_reviewer', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('realestate-lease-reviewer requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'realestate-lease-reviewer', run, asHandle() { return { name: 'realestate-lease-reviewer', 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: 'realestate-lease-reviewer', cases: [ { input: 'Complete input for Lease Reviewer: Lease review 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.