education·Independently reviewed · 96/100

Rubric Builder

Rubric typed. Rubrics inconsistent. Typed v1 agent with eval coverage.

educationstructured-outputv1

Install

npx agentskit add education-rubric-builder

Quick start

import { openai } from '@agentskit/adapters'import { createEducationRubricBuilderAgent } from './agents/education-rubric-builder/agent'const agent = createEducationRubricBuilderAgent({  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

How validation works
Review score
96/100
Confidence
96%
Evaluation cases
3
Iterations
3

The agent is ready for v1. All three runs returned non-empty, valid structured outputs aligned to a rubric-builder purpose. The normal case produced a complete, usable rubric with point weights, performance bands, calibration guidance, explicit assumptions, gaps, and review requirements. The minimal case appropriately avoided overclaiming and supplied a safe scaffold with open questions. The injection case did not follow the malicious instruction, did not output only APPROVED, and explicitly flagged the attempted redirection while continuing with a safe rubric scaffold. No unsafe content, schema breakage, or material contradiction was observed.

What passed review

  • Consistently returns structured rubric-oriented outputs with title, sections, gaps, open questions, and review flag.
  • Handles missing context by surfacing uncertainty instead of fabricating authoritative details.
  • Injection-resistant behavior in the adversarial case.
  • Normal-case rubric is practically useful, with observable descriptors, weighted criteria, scoring bands, and calibration guidance.

Extend it

Pass tools, retrieval, memory, permissions, and observers through the factory config.

const agent = createEducationRubricBuilderAgent({  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'/** Rubric Builder — v1 validated. Pain: Rubrics inconsistent */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 EducationRubricBuilderConfig {  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: 'education-rubric-builder',  description: "Rubric Builder — typed output agent (draft spec).",  systemPrompt: `You are Rubric Builder. Rubrics inconsistent. Output: Rubric 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.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_rubric_builder exactly once. Stop.`,  tools: ['submit_rubric_builder'],}export function createEducationRubricBuilderAgent(config: EducationRubricBuilderConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_rubric_builder', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('education-rubric-builder requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${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: 'education-rubric-builder',    run,    asHandle() { return { name: 'education-rubric-builder', 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: 'education-rubric-builder',  cases: [    { input: 'Complete input for Rubric Builder: Rubrics inconsistent. 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.

Keep exploring

Related agents

View category