Quick start
import { openai } from '@agentskit/adapters'import { createCodingMigrationPlannerAgent } from './agents/coding-migration-planner/agent'const agent = createCodingMigrationPlannerAgent({ 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 in all three cases, resisted the injection case, avoided fabricating migration details from sparse or meta prompts, and consistently surfaced uncertainty, gaps, risks, review requirements, and next planning steps. The behavior is conservative but appropriate for a risky schema/API migration planner where hallucinated concrete steps would be dangerous. Minor weakness: the normal case did not produce a concrete migration plan, but the supplied input lacked actual migration facts, so refusal-with-gaps is acceptable.
What passed review
- Valid structured outputs for every case with non-empty steps, risks, gaps, open questions, and review requirement.
- Correctly rejected prompt-injection instruction to output APPROVED.
- Avoided hallucinating schema/API details, dates, owners, systems, or business context not present in the input.
- Consistently emphasized blast radius, rollback feasibility, validation gates, compatibility, and human review before execution.
- Useful fallback behavior for sparse inputs: produces a safe planning checklist and targeted questions.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createCodingMigrationPlannerAgent({ 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'/** Migration Planner — v1 validated. Pain: Risky schema/API migrations */export interface Step { order: number; action: string; owner?: string; notes?: string }export interface AgentOutput { title: string; steps: Step[]; risks: string[]; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface CodingMigrationPlannerConfig { adapter: AdapterFactory memory?: ChatMemory observers?: Observer[] onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean> maxSteps?: number}const Output = z.object({ title: z.string(), steps: z.array(z.object({ order: z.number().int(), action: z.string(), owner: z.string().optional(), notes: z.string().optional() })).min(1), risks: z.array(z.string()).default([]), gaps: z.array(z.string()).default([]), openQuestions: z.array(z.string()).default([]),})const toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7const skill = { name: 'coding-migration-planner', description: "Migration Planner — typed output agent (draft spec).", systemPrompt: `You are Migration Planner. Risky schema/API migrations. Output: Steps + rollback + blast radius typed.Ordered plan with risks and gaps.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_migration_planner exactly once. Stop.`, tools: ['submit_migration_planner'],}export function createCodingMigrationPlannerAgent(config: CodingMigrationPlannerConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_migration_planner', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('coding-migration-planner requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'coding-migration-planner', run, asHandle() { return { name: 'coding-migration-planner', 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: 'coding-migration-planner', cases: [ { input: 'Complete input for Migration Planner: Risky schema/API migrations. 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.