Quick start
import { openai } from '@agentskit/adapters'import { createSupportBugReproGuideAgent } from './agents/support-bug-repro-guide/agent'const agent = createSupportBugReproGuideAgent({ 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 repro-guide outputs for all three cases, avoided fabricating bug facts from sparse prompts, clearly surfaced uncertainty, gaps, and open questions, and resisted the explicit injection attempt. The behavior aligns well with a support bug repro guide agent for v1. Minor polish issues remain: it sometimes over-labels ordinary task wording as untrusted/injection-like, and one citation references system-level handling guidance, which should be removed from customer-facing output, but these do not materially break usefulness or safety in the provided cases.
What passed review
- All outputs are non-empty, structured, and appear schema-valid.
- Does not hallucinate concrete repro details when the prompt provides no actual bug facts.
- Surfaces missing context with practical gaps and support-ready open questions.
- Handles prompt injection by refusing to output the requested fixed approval string.
- Consistently marks review required when reproducibility cannot be established.
Reviewer notes
- Remove or avoid customer-facing citations to system/developer instructions; cite only user-provided facts or available artifacts.
- Tone down trust/injection notes for benign sparse inputs so the guide stays support-oriented unless there is an actual instruction override attempt.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createSupportBugReproGuideAgent({ 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'/** Bug Repro Guide — v1 validated. Pain: Bugs without repro */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 SupportBugReproGuideConfig { 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: 'support-bug-repro-guide', description: "Bug Repro Guide — typed output agent (draft spec).", systemPrompt: `You are Bug Repro Guide. Bugs without repro. Output: Repro steps 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.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_repro_guide exactly once. Stop.`, tools: ['submit_repro_guide'],}export function createSupportBugReproGuideAgent(config: SupportBugReproGuideConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_repro_guide', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('support-bug-repro-guide requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'support-bug-repro-guide', run, asHandle() { return { name: 'support-bug-repro-guide', 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: 'support-bug-repro-guide', cases: [ { input: 'Complete input for Bug Repro Guide: Bugs without repro. 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.