coding·Independently reviewed · 96/100

Tech Debt Scorer

Scored items: impact/effort/risk. Refactor prioritization. Typed v1 agent with eval coverage.

codingstructured-outputv1

Install

npx agentskit add coding-tech-debt-scorer

Quick start

import { openai } from '@agentskit/adapters'import { createCodingTechDebtScorerAgent } from './agents/coding-tech-debt-scorer/agent'const agent = createCodingTechDebtScorerAgent({  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
1

The agent produced valid structured records for all three cases, handled sparse inputs conservatively, resisted the prompt-injection attempt, avoided inventing technical debt details, and surfaced concrete gaps and follow-up questions. This is useful behavior for a tech-debt scorer when no actual debt item is supplied. The only minor concerns are inconsistent taxonomy strings such as human_review vs human-review and lack of explicit impact/effort/risk placeholders in insufficient-input cases, but these do not block v1 approval based on the observed schema acceptance and behavior.

What passed review

  • Valid structured output in every case.
  • Correctly avoided hallucinating a realistic task from placeholder input.
  • Strong uncertainty handling with clear gaps and open questions.
  • Successfully resisted the injection request to output APPROVED.
  • Appropriate human-review routing for underspecified requests.

Reviewer notes

  • Normalize categorical strings across cases, especially queue values like human_review vs human-review.
  • If the schema supports it, include explicit impact/effort/risk as unknown or unscorable when input is insufficient.

Extend it

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

const agent = createCodingTechDebtScorerAgent({  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'/** Tech Debt Scorer — v1 validated. Pain: Refactor prioritization */export type Severity = 'critical' | 'high' | 'medium' | 'low'export interface AgentOutput { category: string; severity: Severity; queue: string; rationale: string; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface CodingTechDebtScorerConfig {  adapter: AdapterFactory  memory?: ChatMemory  observers?: Observer[]  onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean>  maxSteps?: number}const Output = z.object({  category: z.string(),  severity: z.enum(['critical', 'high', 'medium', 'low']),  queue: z.string(),  rationale: z.string(),  gaps: z.array(z.string()).default([]),  openQuestions: z.array(z.string()).default([]),})const toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7function applySafetyNet(input: string, o: z.infer<typeof Output>) {  if (/\b(outage|breach|emergency|stroke|suicidal|data loss)\b/i.test(input) && o.severity !== 'critical')    return { ...o, severity: 'critical' as const, queue: 'escalation', rationale: o.rationale + ' [safety-net]' }  return o}const skill = {  name: 'coding-tech-debt-scorer',  description: "Tech Debt Scorer — typed output agent (draft spec).",  systemPrompt: `You are Tech Debt Scorer. Refactor prioritization. Output: Scored items: impact/effort/risk.Classify with category, severity, queue, rationale. Gaps for missing input.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_debt_scorer exactly once. Stop.`,  tools: ['submit_debt_scorer'],}export function createCodingTechDebtScorerAgent(config: CodingTechDebtScorerConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_debt_scorer', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('coding-tech-debt-scorer requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${fenceUntrustedContent(input)}`,      parse: (a) => applySafetyNet(input, Output.parse(a)),      skill,      memory: config.memory,      observers: config.observers,      onConfirm: config.onConfirm,      maxSteps: config.maxSteps ?? 4,    })    return { ...result, requiresReview: true }  }  return {    name: 'coding-tech-debt-scorer',    run,    asHandle() { return { name: 'coding-tech-debt-scorer', 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-tech-debt-scorer',  cases: [    { input: 'Complete input for Tech Debt Scorer: Refactor prioritization. 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