Azure AI Foundry connectors

Integrate Azure AI Foundry (Azure OpenAI) models into Bonita processes with enterprise-grade security, Azure AD integration and VNet support.

The Azure AI Foundry connector is part of the Bonita AI Connectors family.

Getting started

Import the bonita-connector-ai-azure module as an extension dependency in your Bonita project. See the AI connectors overview for general setup instructions.

Connection configuration

Azure AI Foundry requires setting the url parameter to your Azure endpoint and the chatModelName to your deployment name. Both are mandatory for this provider.
Parameter Required Description Default

API Key

Yes

Azure OpenAI API key from Azure Portal > Azure AI Foundry > Keys and Endpoint

Resolved from env var AI_API_KEY if not set

Endpoint URL

Yes

Your Azure OpenAI endpoint (e.g. https://my-resource.openai.azure.com)

None — must be set

Model Name (Deployment Name)

Yes

Your Azure deployment name (e.g. gpt-4o)

None — must be set

Temperature

No

Controls randomness (0.0 to 1.0)

Timeout

No

Request timeout in milliseconds

To find your endpoint and key:

  1. Go to the Azure Portal

  2. Navigate to Azure AI Foundry (or Azure OpenAI Service)

  3. Select your resource

  4. Go to Keys and Endpoint

Available models

Azure AI Foundry supports any model deployed in your Azure subscription. Common models include:

  • gpt-4o — Most capable OpenAI model

  • gpt-4o-mini — Cost-effective for simpler tasks

  • gpt-4-turbo — Previous generation

The model name in the connector must match your deployment name in Azure, not the base model name.

Operations

Ask

Send a user prompt (with optional system prompt and documents) to the model and return the generated response.

Parameter Required Description Default

User Prompt

Yes

The prompt to send to the model

System Prompt

No

System instructions to guide the model behavior

You are a polite assistant.

Output JSON Schema

No

JSON Schema to structure the response as JSON

Source Document Reference

No

Bonita process document to include as context

Source Document References

No

List of Bonita process documents to include as context
Parameter Type Description

output

String

The generated response from the model

Classify

Classify a document into one of the predefined categories.

Parameter Required Description Default

Categories

Yes

Comma-separated list of classification categories

Source Document Reference

Yes

Bonita process document to classify

Source Document References

No

List of documents to classify
Parameter Type Description

output

String

JSON with category and confidence fields
Sample classification result
{
  "category": "EXPENSE_REPORT",
  "confidence": 0.89
}

Extract

Extract structured data from a document using field names or a JSON Schema.

Parameter Required Description Default

Fields to Extract

No

Comma-separated list of field names to extract

Output JSON Schema

No

JSON Schema defining the extraction structure

Source Document Reference

Yes

Bonita process document to extract from

Source Document References

No

List of documents to extract from
You must provide at least one of fieldsToExtract or outputJsonSchema parameters.
Parameter Type Description

output

String

JSON with extracted fields

Use cases

Enterprise document processing

Process internal HR and finance documents using Azure AI within your corporate network, leveraging Azure AD authentication and VNet integration for maximum security.

Process flow:

  1. An expense report is submitted through a Bonita form

  2. A service task uses the Extract connector to parse the document

  3. The data is validated against company policies via business rules

  4. A human task presents the extracted data for manager approval

Configuration:

{
  "apiKey": "${AZURE_OPENAI_API_KEY}",
  "url": "https://my-company.openai.azure.com",
  "chatModelName": "gpt-4o",
  "fieldsToExtract": "employeeName,department,expenseDate,totalAmount,currency,category,receipts",
  "outputJsonSchema": "{\"type\":\"object\",\"required\":[\"employeeName\",\"department\",\"expenseDate\",\"totalAmount\",\"currency\",\"category\"],\"properties\":{\"employeeName\":{\"type\":\"string\"},\"department\":{\"type\":\"string\"},\"expenseDate\":{\"type\":\"string\"},\"totalAmount\":{\"type\":\"number\"},\"currency\":{\"type\":\"string\"},\"category\":{\"type\":\"string\"},\"receipts\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"required\":[\"vendor\",\"amount\",\"date\"],\"properties\":{\"vendor\":{\"type\":\"string\"},\"amount\":{\"type\":\"number\"},\"date\":{\"type\":\"string\"}}}}}}"
}

Expected output:

{
  "employeeName": "Sophie Bernard",
  "department": "Sales",
  "expenseDate": "2026-03-15",
  "totalAmount": 847.50,
  "currency": "EUR",
  "category": "Business Travel",
  "receipts": [
    { "vendor": "Air France", "amount": 450.00, "date": "2026-03-10" },
    { "vendor": "Hotel Mercure", "amount": 320.00, "date": "2026-03-11" },
    { "vendor": "Restaurant Le Petit", "amount": 77.50, "date": "2026-03-11" }
  ]
}

CV screening and candidate evaluation

Use Azure AI to evaluate candidate CVs against job descriptions, scoring candidates and extracting key qualifications within the secure corporate network.

Process flow:

  1. A CV document is received through the recruitment process

  2. A service task uses the Ask connector to evaluate the CV against the job requirements

  3. The structured evaluation is stored as BDM objects

  4. A human task presents the evaluation for recruiter review

Configuration:

{
  "apiKey": "${AZURE_OPENAI_API_KEY}",
  "url": "https://my-company.openai.azure.com",
  "chatModelName": "gpt-4o",
  "systemPrompt": "You are an HR assistant. Evaluate candidates objectively based on qualifications, experience, and job fit. Be fair and unbiased.",
  "userPrompt": "Evaluate this CV against the following job description and provide a structured assessment.\n\nJob: ${jobTitle}\nRequirements: ${jobRequirements}\n\nProvide a fit score (0-100), key strengths, gaps, and a recommendation.",
  "outputJsonSchema": "{\"type\":\"object\",\"required\":[\"fitScore\",\"strengths\",\"gaps\",\"recommendation\"],\"properties\":{\"fitScore\":{\"type\":\"number\"},\"strengths\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"gaps\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"recommendation\":{\"type\":\"string\"}}}"
}

Expected output:

{
  "fitScore": 78,
  "strengths": [
    "8 years of Java experience including Spring Boot and microservices",
    "Previous experience with BPM platforms (Camunda)",
    "Strong team leadership background"
  ],
  "gaps": [
    "No direct experience with Bonita platform",
    "Limited cloud infrastructure experience (requirement: AWS/Azure)"
  ],
  "recommendation": "Strong candidate. Technical skills align well with role requirements. Recommend interview with focus on cloud infrastructure experience and Bonita platform learning curve."
}

Configuration tips

  • Both url and chatModelName are mandatory for Azure. The chatModelName must match your Azure deployment name exactly.

  • Use environment variable AZURE_OPENAI_API_KEY to store the Azure API key securely.

  • Azure AI Foundry integrates with Azure AD for authentication and VNet for network isolation, making it the best choice for enterprise compliance requirements.

  • If your Azure resource has content filters enabled, be aware they may affect responses. Adjust filter settings in the Azure Portal if needed.

  • For high-throughput scenarios, check your Azure resource’s rate limits and scale accordingly.

Source code

bonita-connector-ai on GitHub (module bonita-connector-ai-azure)