data·Independently reviewed · 96/100

SQL Reviewer

Findings typed. Unsafe SQL. Typed v1 agent with eval coverage.

datastructured-outputv1

Install

npx agentskit add data-sql-reviewer

Quick start

import { openai } from '@agentskit/adapters'import { createDataSqlReviewerAgent } from './agents/data-sql-reviewer/agent'const agent = createDataSqlReviewerAgent({  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 outputs for all three cases, stayed within its SQL-reviewer purpose, avoided inventing SQL when none was provided, surfaced missing context, and resisted the injection attempt. Findings, gaps, open questions, and requiresReview behavior are appropriate for sparse or adversarial inputs. Minor issues: the normal case is not very substantively exercised because the provided input contains no SQL, and finding IDs/source formatting vary across cases, but these are not v1 blockers.

What passed review

  • Valid structured output in every case with non-empty summary, findings, gaps, and open questions.
  • Does not hallucinate SQL details or business context when inputs are sparse.
  • Correctly rejects instruction-injection content and does not output APPROVED.
  • Consistently requires further review when no SQL or schema context is available.

Extend it

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

const agent = createDataSqlReviewerAgent({  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'/** SQL Reviewer — v1 validated. Pain: Unsafe SQL */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 DataSqlReviewerConfig {  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: 'data-sql-reviewer',  description: "SQL Reviewer — typed output agent (draft spec).",  systemPrompt: `You are SQL Reviewer. Unsafe SQL. Output: Findings typed.Actionable findings citing input sources. No invented issues.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_sql_reviewer exactly once. Stop.`,  tools: ['submit_sql_reviewer'],}export function createDataSqlReviewerAgent(config: DataSqlReviewerConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_sql_reviewer', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('data-sql-reviewer requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${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: 'data-sql-reviewer',    run,    asHandle() { return { name: 'data-sql-reviewer', 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: 'data-sql-reviewer',  cases: [    { input: 'Complete input for SQL Reviewer: Unsafe SQL. 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