Quick start
import { openai } from '@agentskit/adapters'import { createCodingPerformanceInterpreterAgent } from './agents/coding-performance-interpreter/agent'const agent = createCodingPerformanceInterpreterAgent({ 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 for all three cases, stayed within its Performance Interpreter purpose, refused to invent bottlenecks without evidence, surfaced concrete gaps and open questions, and handled the injection attempt correctly. Behavior is conservative but useful for sparse inputs. Minor weakness: it sometimes over-labels benign sparse task text as instruction-like untrusted content, but that does not break usefulness or safety.
What passed review
- Consistently avoids hallucinating performance findings when no Lighthouse, bundle, trace, or telemetry data is provided.
- Structured outputs are populated with summary, findings, gaps, openQuestions, and review-needed posture.
- Injection case correctly ignores the request to output APPROVED and treats it as data.
- Recommendations ask for the right artifacts and context for a performance interpretation workflow.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createCodingPerformanceInterpreterAgent({ 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'/** Performance Interpreter — v1 validated. Pain: Lighthouse/bundle reports opaque */export interface Finding { id: string; severity: 'critical' | 'high' | 'medium' | 'low' | 'info'; message: string; source?: string; recommendation?: string }export interface AgentOutput { summary: string; findings: Finding[]; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface CodingPerformanceInterpreterConfig { adapter: AdapterFactory memory?: ChatMemory observers?: Observer[] onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean> maxSteps?: number}const Output = z.object({ summary: z.string(), findings: z.array(z.object({ id: z.string(), severity: z.enum(['critical', 'high', 'medium', 'low', 'info']), message: z.string(), source: z.string().optional(), recommendation: z.string().optional(), })), 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-performance-interpreter', description: "Performance Interpreter — typed output agent (draft spec).", systemPrompt: `You are Performance Interpreter. Lighthouse/bundle reports opaque. Output: Bottlenecks typed.Actionable findings citing input sources. No invented issues.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_performance_interpreter exactly once. Stop.`, tools: ['submit_performance_interpreter'],}export function createCodingPerformanceInterpreterAgent(config: CodingPerformanceInterpreterConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_performance_interpreter', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('coding-performance-interpreter requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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-performance-interpreter', run, asHandle() { return { name: 'coding-performance-interpreter', 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-performance-interpreter', cases: [ { input: 'Complete input for Performance Interpreter: Lighthouse/bundle reports opaque. 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.