JSON extraction

Basic

This example pastes free text and returns a typed person record that matches a Zod schema. The model returns a complete structured response with no intermediate text events.

Live demo: react.fency.ai/json-extraction
GitHub folder: fency-react-examples/app/json-extraction

Prerequisites

You need a publishable key for the React SDK and a secret key for your server routes. This example uses sessions to authenticate the client.

The example uses Clerk for user authentication. That is part of the app shell and not specific to Fency.

Server session routes

The example defines two session routes: one for stream sessions and one for agent task sessions. Each route is standalone so you can read a single file and see the full POST /v1/sessions call.

Create the stream session route at api/create-stream-session/route.ts:

import { NextResponse } from 'next/server'
import { getAuthorizedUserId } from '../../../auth'
import { sessionClientTokenSchema } from '../../sessionClientTokenSchema'

export async function POST() {
  const userId = await getAuthorizedUserId()
  if (!userId) {
    return NextResponse.json({ error: 'Unauthorized' }, { status: 401 })
  }

  const secretKey = process.env.FENCY_SECRET_KEY
  if (!secretKey) {
    throw new Error('FENCY_SECRET_KEY is not defined.')
  }

  const response = await fetch('https://api.fency.ai/v1/sessions', {
    method: 'POST',
    headers: {
      Authorization: `Bearer ${secretKey}`,
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({ createStream: {} }),
  })
  if (!response.ok) {
    throw new Error('Failed to create Fency session.')
  }
  return NextResponse.json(sessionClientTokenSchema.parse(await response.json()))
}

Create the agent task session route at api/create-agent-task-session/route.ts:

import { NextResponse } from 'next/server'
import { getAuthorizedUserId } from '../../../auth'
import { sessionClientTokenSchema } from '../../sessionClientTokenSchema'

export async function POST() {
  const userId = await getAuthorizedUserId()
  if (!userId) {
    return NextResponse.json({ error: 'Unauthorized' }, { status: 401 })
  }

  const secretKey = process.env.FENCY_SECRET_KEY
  if (!secretKey) {
    throw new Error('FENCY_SECRET_KEY is not defined.')
  }

  const response = await fetch('https://api.fency.ai/v1/sessions', {
    method: 'POST',
    headers: {
      Authorization: `Bearer ${secretKey}`,
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      createAgentTask: {
        taskType: 'STRUCTURED_CHAT_COMPLETION',
        metadata: { userId },
      },
    }),
  })
  if (!response.ok) {
    throw new Error('Failed to create Fency session.')
  }
  return NextResponse.json(sessionClientTokenSchema.parse(await response.json()))
}

Extraction schema

Define the extraction schema using Zod in extractionSchema.ts:

import { z } from 'zod'

export const extractionSchema = z.object({
  name: z.string().describe('Full name of the person'),
  role: z.string().describe('Job title or role'),
  company: z.string().describe('Company or organization'),
  email: z
    .string()
    .describe('Email address if mentioned, otherwise an empty string'),
  summary: z.string().describe('One-sentence summary of the person'),
})

export type Extraction = z.infer<typeof extractionSchema>

export const extractionJsonSchema = JSON.stringify(
  z.toJSONSchema(extractionSchema),
)

The schema descriptions guide the model on what to extract for each field.

Client provider and hooks

Both session routes respond with a client token. Parse that response with a Zod schema in sessionClientTokenSchema.ts so a bad shape throws:

import { z } from 'zod'

export const sessionClientTokenSchema = z.object({
  clientToken: z.string(),
})

Load the Fency client and wrap your component tree with FencyProvider in page.tsx:

'use client'

import { loadFency } from '@fencyai/js'
import { FencyProvider } from '@fencyai/react'
import { Extractor } from './components/Extractor'
import { sessionClientTokenSchema } from './sessionClientTokenSchema'

const publishableKey = process.env.NEXT_PUBLIC_FENCY_PUBLISHABLE_KEY
if (!publishableKey) {
  throw new Error('NEXT_PUBLIC_FENCY_PUBLISHABLE_KEY is not defined.')
}

const fency = loadFency({
  publishableKey,
})

async function fetchCreateStreamClientToken() {
  const res = await fetch(
    '/json-extraction/api/create-stream-session',
    { method: 'POST' },
  )
  if (!res.ok) {
    throw new Error('Failed to create stream session')
  }
  const { clientToken } = sessionClientTokenSchema.parse(await res.json())
  return { clientToken }
}

export default function JsonExtractionPage() {
  return (
    <FencyProvider
      fency={fency}
      fetchCreateStreamClientToken={fetchCreateStreamClientToken}
    >
      <Extractor />
    </FencyProvider>
  )
}

Use useAgentTasks and createAgentTask to extract structured data in hooks/useJsonExtraction.ts:

'use client'

import { useAgentTasks } from '@fencyai/react'
import { useState } from 'react'
import {
  extractionJsonSchema,
  extractionSchema,
  type Extraction,
} from '../extractionSchema'
import { sessionClientTokenSchema } from '../sessionClientTokenSchema'

async function fetchCreateAgentTaskClientToken() {
  const res = await fetch(
    '/json-extraction/api/create-agent-task-session',
    { method: 'POST' },
  )
  if (!res.ok) {
    throw new Error('Failed to create agent task session')
  }
  const { clientToken } = sessionClientTokenSchema.parse(await res.json())
  return { clientToken }
}

export function useJsonExtraction() {
  const [isSubmitting, setIsSubmitting] = useState(false)
  const [latestResult, setLatestResult] = useState<Extraction | null>(null)

  const { latest, createAgentTask } = useAgentTasks({})

  async function extract(text: string) {
    setIsSubmitting(true)
    setLatestResult(null)

    try {
      const response = await createAgentTask(
        {
          type: 'StructuredChatCompletion',
          messages: [
            {
              role: 'SYSTEM',
              content:
                'Extract a single person record from the user text. Use empty strings for fields that are not mentioned.',
            },
            { role: 'USER', content: text },
          ],
          model: 'anthropic/claude-sonnet-4.6',
          jsonSchema: extractionJsonSchema,
        },
        { fetchCreateAgentTaskClientToken },
      )

      if (response.type !== 'success') {
        throw new Error(response.error.message)
      }
      if (response.response.taskType !== 'StructuredChatCompletion') {
        throw new Error('Unexpected StructuredChatCompletion outcome.')
      }
      setLatestResult(
        extractionSchema.parse(
          JSON.parse(response.response.response.jsonResponse),
        ),
      )
    } finally {
      setIsSubmitting(false)
    }
  }

  return {
    latestTask: latest,
    latestResult,
    isSubmitting,
    extract,
  }
}

Render progress and the parsed record in components/Extractor.tsx:

'use client'

import { Alert, Badge, Card, Text, Title } from '@mantine/core'
import { AgentTaskProgress } from '@fencyai/react'
import { useJsonExtraction } from '../hooks/useJsonExtraction'
import { ExtractionForm } from './ExtractionForm'
import { RecordCard } from './RecordCard'
import { SchemaPreview } from './SchemaPreview'

export function Extractor() {
  const { latestTask, latestResult, isSubmitting, extract } =
    useJsonExtraction()

  return (
    <div className="mx-auto flex w-full max-w-5xl flex-col gap-6 px-4 py-6">
      <div>
        <Badge size="sm" variant="light" color="green" mb={4}>
          Basic
        </Badge>
        <Title order={1} size="h4">
          JSON extraction
        </Title>
        <Text size="sm" c="dimmed">
          Paste free text and get a typed record back.
        </Text>
      </div>

      <Card withBorder padding="lg" radius="md">
        <Title order={2} size="h5" mb="sm">
          Schema
        </Title>
        <SchemaPreview />
      </Card>

      <Card withBorder padding="lg" radius="md">
        <Title order={2} size="h5" mb="sm">
          Source text
        </Title>
        <ExtractionForm isSubmitting={isSubmitting} onExtract={extract} />
      </Card>

      {latestTask?.error ? (
        <Alert color="red">{latestTask.error.message}</Alert>
      ) : latestTask ? (
        <AgentTaskProgress agentTask={latestTask} />
      ) : null}

      {latestResult ? (
        <RecordCard title="Latest result" record={latestResult} />
      ) : null}
    </div>
  )
}

The AgentTaskProgress component displays progress while the extraction task runs. The result is parsed with the Zod schema to ensure type safety. components/ExtractionForm.tsx owns the textarea and calls extract.

Running locally

Clone the repository and install dependencies:

git clone https://github.com/fencyai/fency-react-examples.git
cd fency-react-examples
npm install

Copy .env.example to .env.local and add your Fency keys:

cp .env.example .env.local

Set the keys in .env.local:

FENCY_SECRET_KEY=sk_...
NEXT_PUBLIC_FENCY_PUBLISHABLE_KEY=pk_...

Initialize Clerk authentication (provisions a development app):

npx -y clerk@latest init --keyless -y

Start the development server:

npm run dev

Open http://localhost:3000/json-extraction. Sign up from the header, then paste free text to extract structured JSON matching the schema.