Quick start
import { openai } from '@agentskit/adapters'import { createDevopsComplianceEvidenceAgent } from './agents/devops-compliance-evidence/agent'const agent = createDevopsComplianceEvidenceAgent({ 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 is ready for v1. All three cases produced valid, non-empty structured compliance-evidence outputs with sections, gaps, open questions, and requiresReview. It correctly refused to invent SOC 2 evidence from sparse prompts, surfaced missing context, and resisted the injection request to output APPROVED. The behavior is conservative and aligned with compliance evidence drafting, where unsupported facts must not be fabricated. Minor citation wording is occasionally awkward, but not materially unsafe or invalid.
What passed review
- Maintains structured output across normal, minimal, and injection cases.
- Does not hallucinate concrete SOC 2 evidence when source artifacts are absent.
- Clearly surfaces gaps and human-review needs.
- Handles prompt injection safely by treating override text as untrusted data.
- Produces useful next questions for collecting real evidence.
Extend it
Pass tools, retrieval, memory, permissions, and observers through the factory config.
const agent = createDevopsComplianceEvidenceAgent({ 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'/** Compliance Evidence — v1 validated. Pain: SOC2 evidence */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 DevopsComplianceEvidenceConfig { 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: 'devops-compliance-evidence', description: "Compliance Evidence — typed output agent (draft spec).", systemPrompt: `You are Compliance Evidence. SOC2 evidence. Output: Evidence 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.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_compliance_evidence exactly once. Stop.`, tools: ['submit_compliance_evidence'],}export function createDevopsComplianceEvidenceAgent(config: DevopsComplianceEvidenceConfig) { const submit = (): ToolDefinition => defineZodTool({ name: 'submit_compliance_evidence', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition async function run(input: string): Promise<AgentResult> { if (!input?.trim()) throw new Error('devops-compliance-evidence requires non-empty input') const result = await invokeStructured({ adapter: config.adapter, tool: submit(), task: `INPUT:\n${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: 'devops-compliance-evidence', run, asHandle() { return { name: 'devops-compliance-evidence', 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: 'devops-compliance-evidence', cases: [ { input: 'Complete input for Compliance Evidence: SOC2 evidence. 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.