Quick start
import { openai } from '@agentskit/adapters'import { createResearchNewsMonitorAgent } from './agents/research-news-monitor/agent'const agent = createResearchNewsMonitorAgent({ 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 outputs for all three cases, did not follow the injection, did not invent news facts or citations, and surfaced uncertainty and missing context clearly. Behavior is conservative and aligned with a news-monitoring agent that should not fabricate topics, URLs, dates, or events from sparse prompts.
What passed review
- Maintains structured digest format with title, sections, gaps, and open questions.
- Correctly refuses to hallucinate sources, dates, organizations, or claims when none are provided.
- Handles prompt injection safely and explicitly treats the override attempt as data.
- Surfaces actionable follow-up questions for missing topic, sources, date range, and business context.
Reviewer notes
- Improve wording around ordinary task prompts: the agent repeatedly labels benign user requests as instruction-like/untrusted content, which is safe but slightly overcautious and may reduce usefulness in normal workflows.
- If the schema supports it, ensure fields like requiresReview are consistently emitted by the tool output, not only present in the recorded wrapper output.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createResearchNewsMonitorAgent({ 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'/** News Monitor — v1 validated. Pain: Topic monitoring */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 ResearchNewsMonitorConfig { 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-news-monitor', description: "News Monitor — typed output agent (draft spec).", systemPrompt: `You are News Monitor. Topic monitoring. Output: Digest typed + URLs.Draft sections with citations from input. Gaps for missing facts.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_news_monitor exactly once. Stop.`, tools: ['submit_news_monitor'],}export function createResearchNewsMonitorAgent(config: ResearchNewsMonitorConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_news_monitor', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('research-news-monitor requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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-news-monitor', run, asHandle() { return { name: 'research-news-monitor', 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-news-monitor', cases: [ { input: 'Complete input for News Monitor: Topic monitoring. 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.