Skip to main content

​
Install

pip install synap-haystack

​
What’s included

ClassPurpose
SynapRetrieverPipeline component that fetches Synap memories as Document objects
SynapMemoryWriterPipeline component that writes conversation turns back to Synap

​
SynapRetriever

Retrieves memories as Haystack Document objects for use in RAG pipelines. 
from haystack import Pipeline
from haystack.components.builders import PromptBuilder
from synap_haystack import SynapRetriever

retriever = SynapRetriever(
    sdk=sdk,
    user_id="alice",
    customer_id="acme",   # optional
    max_results=6,
    mode="fast",          # "fast" or "accurate"
)

pipeline = Pipeline()
pipeline.add_component("retriever", retriever)
pipeline.add_component("prompt_builder", PromptBuilder(template=your_template))
pipeline.connect("retriever.documents", "prompt_builder.documents")

result = pipeline.run({"retriever": {"query": "project deadlines"}})
Each returned Document has:
  • content — the memory text
  • meta["type"] — memory type (e.g. "fact", "preference")
  • meta["confidence"] — relevance score

​
SynapMemoryWriter

Records conversation turns back into Synap. Place it at the end of your pipeline after the LLM response. 
from synap_haystack import SynapRetriever, SynapMemoryWriter

writer = SynapMemoryWriter(
    sdk=sdk,
    conversation_id="conv-001",
    user_id="alice",
    customer_id="acme",
)

pipeline = Pipeline()
pipeline.add_component("retriever", retriever)
pipeline.add_component("llm", your_llm)
pipeline.add_component("writer", writer)

pipeline.connect("retriever.documents", "llm.documents")
pipeline.connect("llm.replies", "writer.replies")
SynapMemoryWriter surfaces write failures explicitly (raises SynapIntegrationError) so your pipeline knows if memory persistence failed.

​
Full pipeline example

from haystack import Pipeline
from haystack.components.builders import PromptBuilder
from haystack.components.generators import OpenAIGenerator
from synap_haystack import SynapRetriever, SynapMemoryWriter

retriever = SynapRetriever(sdk=sdk, user_id="alice")
writer = SynapMemoryWriter(sdk=sdk, conversation_id="conv-001", user_id="alice")

template = """
Given this context about the user:
{% for doc in documents %}
- {{ doc.content }}
{% endfor %}
Answer: {{ query }}
"""

pipeline = Pipeline()
pipeline.add_component("retriever", retriever)
pipeline.add_component("prompt", PromptBuilder(template=template))
pipeline.add_component("llm", OpenAIGenerator(model="gpt-4o"))
pipeline.add_component("writer", writer)

pipeline.connect("retriever.documents", "prompt.documents")
pipeline.connect("prompt.prompt", "llm.prompt")
pipeline.connect("llm.replies", "writer.replies")

result = pipeline.run({
    "retriever": {"query": "What are my priorities?"},
    "prompt": {"query": "What are my priorities?"},
})

​
Next steps

LangChain

Memory and retriever for LangChain.

LlamaIndex

BaseMemory and retriever for LlamaIndex.