Skip to main content
For the complete documentation index for agents and LLMs, see llms.txt.

SuperComponent

Wrap an entire pipeline in SuperComponent and use it as a single reusable component with simplified inputs and outputs.

Key Features​

  • Encapsulates a complete pipeline as a single reusable component.
  • Supports custom input and output name remapping via input_mapping and output_mapping.
  • Simplifies complex pipelines by hiding internal complexity.
  • Useful for defining tools for agents that execute entire pipelines.
  • Inputs and outputs are dynamically determined by the input_mapping and output_mapping configurations.

Configuration​

  1. Drag the SuperComponent component onto the canvas from the Component Library.
  2. Click on the component to open the configuration panel.
  3. On the General tab:
    • Configure the wrapped pipeline by defining all its components and connections in the YAML.
    • Set input_mapping to map external input names to pipeline input socket paths.
    • Set output_mapping to map pipeline output socket paths to external output names.

Connections​

Inputs and outputs are dynamically defined based on the input_mapping and output_mapping configurations. By default, all pipeline inputs are exposed as inputs and all pipeline outputs are exposed as outputs.

Source Code​

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

Usage Examples​

Basic Configuration​

  rag_tool:
    type: haystack.core.super_component.super_component.SuperComponent
    init_parameters:
      input_mapping:
        query:
        - bm25_retriever.query
        - query_embedder.text
        - ranker.query
        - prompt_builder.query
      output_mapping:
        generator.replies: answer
        ranker.documents: documents
      pipeline:
        components:
          bm25_retriever:
            type: haystack_integrations.components.retrievers.opensearch.bm25_retriever.OpenSearchBM25Retriever
            init_parameters:
              document_store:
                type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
                init_parameters:
                  index: default
                  max_chunk_bytes: 104857600
                  embedding_dim: 768
                  return_embedding: false
                  create_index: true
              top_k: 20
              fuzziness: 0
          query_embedder:
            type: deepset_cloud_custom_nodes.embedders.nvidia.text_embedder.DeepsetNvidiaTextEmbedder
            init_parameters:
              normalize_embeddings: true
              model: intfloat/e5-base-v2
          embedding_retriever:
            type: haystack_integrations.components.retrievers.opensearch.embedding_retriever.OpenSearchEmbeddingRetriever
            init_parameters:
              document_store:
                type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
                init_parameters:
                  index: default
                  max_chunk_bytes: 104857600
                  embedding_dim: 768
                  return_embedding: false
                  create_index: true
              top_k: 20
          document_joiner:
            type: haystack.components.joiners.document_joiner.DocumentJoiner
            init_parameters:
              join_mode: concatenate
          ranker:
            type: deepset_cloud_custom_nodes.rankers.nvidia.ranker.DeepsetNvidiaRanker
            init_parameters:
              model: intfloat/simlm-msmarco-reranker
              top_k: 8
          prompt_builder:
            type: haystack.components.builders.prompt_builder.PromptBuilder
            init_parameters:
              template: "You are a helpful assistant answering the user's questions based on the provided documents.\nDo not
                use your own knowledge.\n\nProvided documents:\n{% for document in documents %}\nDocument [{{ loop.index }}]:\n\
                {{ document.content }}\n{% endfor %}\n\nQuestion: {{ query }}\nAnswer:"
          generator:
            type: haystack.components.generators.chat.openai.OpenAIChatGenerator
            init_parameters:
              api_key:
                type: env_var
                env_vars:
                - OPENAI_API_KEY
                strict: false
              model: gpt-4o
              generation_kwargs:
                max_tokens: 1000
                temperature: 0.7
        connections:
        - sender: bm25_retriever.documents
          receiver: document_joiner.documents
        - sender: query_embedder.embedding
          receiver: embedding_retriever.query_embedding
        - sender: embedding_retriever.documents
          receiver: document_joiner.documents
        - sender: document_joiner.documents
          receiver: ranker.documents
        - sender: ranker.documents
          receiver: prompt_builder.documents
        - sender: prompt_builder.prompt
          receiver: generator.messages

This is an example RAG pipeline using SuperComponent to wrap a retrieval sub-pipeline as a reusable tool:

components:
  rag_tool:
    type: haystack.core.super_component.super_component.SuperComponent
    init_parameters:
      input_mapping:
        query:
        - bm25_retriever.query
        - query_embedder.text
        - ranker.query
        - prompt_builder.query
      output_mapping:
        generator.replies: answer
        ranker.documents: documents
      pipeline:
        components:
          bm25_retriever:
            type: haystack_integrations.components.retrievers.opensearch.bm25_retriever.OpenSearchBM25Retriever
            init_parameters:
              document_store:
                type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
                init_parameters:
                  hosts:
                  index: 'default'
                  max_chunk_bytes: 104857600
                  embedding_dim: 768
                  return_embedding: false
                  method:
                  mappings:
                  settings:
                  create_index: true
                  http_auth:
                  use_ssl:
                  verify_certs:
                  timeout:
              top_k: 20
              fuzziness: 0
          query_embedder:
            type: deepset_cloud_custom_nodes.embedders.nvidia.text_embedder.DeepsetNvidiaTextEmbedder
            init_parameters:
              normalize_embeddings: true
              model: intfloat/e5-base-v2
          embedding_retriever:
            type: haystack_integrations.components.retrievers.opensearch.embedding_retriever.OpenSearchEmbeddingRetriever
            init_parameters:
              document_store:
                type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
                init_parameters:
                  hosts:
                  index: 'default'
                  max_chunk_bytes: 104857600
                  embedding_dim: 768
                  return_embedding: false
                  method:
                  mappings:
                  settings:
                  create_index: true
                  http_auth:
                  use_ssl:
                  verify_certs:
                  timeout:
              top_k: 20
          document_joiner:
            type: haystack.components.joiners.document_joiner.DocumentJoiner
            init_parameters:
              join_mode: concatenate
          ranker:
            type: deepset_cloud_custom_nodes.rankers.nvidia.ranker.DeepsetNvidiaRanker
            init_parameters:
              model: intfloat/simlm-msmarco-reranker
              top_k: 8
          prompt_builder:
            type: haystack.components.builders.prompt_builder.PromptBuilder
            init_parameters:
              template: "You are a helpful assistant answering the user's questions based on the provided documents.\nDo not use your own knowledge.\n\nProvided documents:\n{% for document in documents %}\nDocument [{{ loop.index }}]:\n{{ document.content }}\n{% endfor %}\n\nQuestion: {{ query }}\nAnswer:"
          generator:
            type: haystack.components.generators.chat.openai.OpenAIChatGenerator
            init_parameters:
              api_key:
                type: env_var
                env_vars:
                - OPENAI_API_KEY
                strict: false
              model: gpt-4o
              generation_kwargs:
                max_tokens: 1000
                temperature: 0.7
        connections:
        - sender: bm25_retriever.documents
          receiver: document_joiner.documents
        - sender: query_embedder.embedding
          receiver: embedding_retriever.query_embedding
        - sender: embedding_retriever.documents
          receiver: document_joiner.documents
        - sender: document_joiner.documents
          receiver: ranker.documents
        - sender: ranker.documents
          receiver: prompt_builder.documents
        - sender: prompt_builder.prompt
          receiver: generator.messages

  answer_builder:
    type: deepset_cloud_custom_nodes.augmenters.deepset_answer_builder.DeepsetAnswerBuilder
    init_parameters:
      reference_pattern: acm

connections:
- sender: rag_tool.answer
  receiver: answer_builder.replies
- sender: rag_tool.documents
  receiver: answer_builder.documents

inputs:
  query:
  - "rag_tool.query"
  - "answer_builder.query"

outputs:
  documents: "rag_tool.documents"
  answers: "answer_builder.answers"

max_runs_per_component: 100

metadata: {}

Parameters​

Inputs​

Inputs are dynamically defined based on the input_mapping configuration. By default, all pipeline inputs are exposed.

Outputs​

Outputs are dynamically defined based on the output_mapping configuration. By default, all pipeline outputs are exposed.

Init Parameters​

These are the parameters you can configure in Pipeline Builder:

ParameterTypeDefaultDescription
pipelineUnion[Pipeline, AsyncPipeline]The pipeline instance to be wrapped.
input_mappingOptional[Dict[str, List[str]]]NoneA dictionary mapping component input names to pipeline input socket paths. If not provided, a default mapping is created based on all pipeline inputs. Example: {"query": ["retriever.query", "prompt_builder.query"]}
output_mappingOptional[Dict[str, str]]NoneA dictionary mapping pipeline output socket paths to component output names. If not provided, a default mapping is created based on all pipeline outputs. Example: {"generator.replies": "replies"}

Run Method Parameters​

The run method parameters are dynamically determined by the input_mapping configuration. Each key in the input mapping becomes a parameter that can be passed at runtime.