OpenAPIServiceConnector

Invoke OpenAPI services from tool calls in chat messages. Use this component in agent pipelines where an LLM selects tools and this component executes them against a REST API.

Key Features

• Processes ChatMessage objects with tool calls and invokes the matching OpenAPI operations.
• Authenticates with http and apiKey security schemes defined in the OpenAPI spec.
• Returns service responses as ChatMessage objects for downstream LLM components.
• Works with function definitions produced by OpenAPIServiceToFunctions.

Configuration

Add Workspace-Level Integration

1. Click your profile icon and choose Settings.
2. Go to Workspace>Integrations.
3. Find the provider you want to connect and click Connect next to them.
4. Enter the API key and any other required details.
5. Click Connect. You can use this integration in pipelines and indexes in the current workspace.

Add Organization-Level Integration

1. Click your profile icon and choose Settings.
2. Go to Organization>Integrations.
3. Find the provider you want to connect and click Connect next to them.
4. Enter the API key and any other required details.
5. Click Connect. You can use this integration in pipelines and indexes in all workspaces in the current organization.

1. Drag the OpenAPIServiceConnector component onto the canvas from the Component Library.
2. Connect an upstream component that provides OpenAPI function definitions (typically OpenAPIServiceToFunctions).
3. Connect chat messages with tool calls to the messages input.
4. Pass the resolved OpenAPI spec and credentials at runtime through service_openapi_spec and service_credentials.

Connections

OpenAPIServiceConnector receives a list of ChatMessage objects (the last message must be from the assistant and contain tool calls), an OpenAPI specification, and optional credentials. It outputs service_response as a list of ChatMessage objects with JSON responses from the service.

Connect it downstream from a chat generator that produces tool calls and upstream from components that consume tool results.

Source Code

To check this component's source code, open openapi_service.py in the Haystack Core Integrations repository.

Usage Examples

Basic Configuration

  OpenAPIServiceConnector:
    type: haystack_integrations.components.connectors.openapi.openapi_service.OpenAPIServiceConnector
    init_parameters: {}

In an agent pipeline, convert OpenAPI specs first, then pass a single resolved spec into the connector with OutputAdapter. The converter outputs a list of specs; the connector expects one dict:

# haystack-pipeline
components:
  spec_converter:
    type: haystack_integrations.components.converters.openapi.openapi_functions.OpenAPIServiceToFunctions
    init_parameters: {}

  openapi_spec_adapter:
    type: haystack.components.converters.output_adapter.OutputAdapter
    init_parameters:
      template: "{{ specs[0] }}"
      output_type: Dict[str, Any]
      unsafe: true

  service_connector:
    type: haystack_integrations.components.connectors.openapi.openapi_service.OpenAPIServiceConnector
    init_parameters: {}

connections:
- sender: spec_converter.openapi_specs
  receiver: openapi_spec_adapter.specs
- sender: openapi_spec_adapter.output
  receiver: service_connector.service_openapi_spec

inputs:
  sources:
  - spec_converter.sources
  messages:
  - service_connector.messages

outputs:
  service_response: service_connector.service_response

Parameters

Inputs

Parameter	Type	Description
messages	List[ChatMessage]	Chat messages to process. The last message must be from the assistant and contain tool calls.
service_openapi_spec	Dict[str, Any]	The OpenAPI JSON specification with resolved references.
service_credentials	Dict | str | None	Credentials for authentication. A string for single-scheme APIs, or a dictionary keyed by security scheme name.

Outputs

Parameter	Type	Description
service_response	List[ChatMessage]	Responses from the service as user-role chat messages with JSON content.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
ssl_verify	bool | str | None	None	Whether to verify SSL for requests. Pass a string to use it as the CA certificate.

Run Method Parameters

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

Parameter	Type	Default	Description
messages	List[ChatMessage]		Chat messages with tool calls to execute.
service_openapi_spec	Dict[str, Any]		The OpenAPI specification of the service.
service_credentials	Dict | str | None	None	Credentials for service authentication.

Related Information

• OpenAPIServiceToFunctions
• OpenAPIConnector