Pipelines | Imaginist

📄️ EmailParserPipeline

Converts email objects into structured documents for downstream processing.

Parses documents into structured text for downstream pipelines.

Cleans and splits documents into smaller chunks for processing.

Converts documents into vector embeddings for search and retrieval.

Writes structured or unstructured content into a selected vector store index with metadata validation.

Fetches relevant documents for a query (RAG).

Routes and processes emails end-to-end.

Routes documents based on metadata conditions.

Classifies documents into predefined categories.

Extracts structured data from documents using configurable labels and optional grouping.

Convert natural language questions to SQL and execute them on a database.

Calls OpenAPI endpoints and returns responses.

Searches a vector store for similar items.

Mask sensitive information in documents using the Presidio Masker Guardrail.