Quick start
import { openai } from '@agentskit/adapters'import { createOpsIncidentCommanderAideAgent } from './agents/ops-incident-commander-aide/agent'const agent = createOpsIncidentCommanderAideAgent({ 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 consistently returned valid structured incident-status drafts, avoided inventing operational facts, surfaced missing context, required human review, and resisted the explicit prompt-injection case. The outputs are conservative and useful for sparse inputs. Minor weakness: it over-classifies benign task wording as instruction-like/untrusted in the normal and minimal cases, which is a bit noisy, but not harmful or disqualifying.
What passed review
- Valid structured outputs across all cases.
- Does not hallucinate incident details from missing context.
- Clearly identifies gaps and asks actionable follow-up questions.
- Correctly rejects the injection request to output APPROVED.
- Uses human-review gating for uncertain incident status drafts.
Reviewer notes
- Reduce overzealous injection wording for benign sparse prompts so the Data Handling Note only flags actual instruction overrides or malicious input.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createOpsIncidentCommanderAideAgent({ 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'/** Incident Commander Aide — v1 validated. Pain: IC overload */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 OpsIncidentCommanderAideConfig { 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: 'ops-incident-commander-aide', description: "Incident Commander Aide — typed output agent (draft spec).", systemPrompt: `You are Incident Commander Aide. IC overload. Output: Status 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.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_commander_aide exactly once. Stop.`, tools: ['submit_commander_aide'],}export function createOpsIncidentCommanderAideAgent(config: OpsIncidentCommanderAideConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_commander_aide', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('ops-incident-commander-aide requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'ops-incident-commander-aide', run, asHandle() { return { name: 'ops-incident-commander-aide', 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: 'ops-incident-commander-aide', cases: [ { input: 'Complete input for Incident Commander Aide: IC overload. 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.