Document Processor integration

What the Document Processor package does

Parse and process documents for RAG applications. Supports PDF, DOCX, PPTX, TXT, HTML, Markdown and more. Extract text, chunk content for embeddings.

Parse and process documents for RAG pipelines. Powered by the unstructured library, it supports PDF, DOCX, PPTX, TXT, HTML, Markdown, EML, RTF, EPUB, ODT, CSV, TSV, JSON, and XML. Provides a full document-to-embeddings pipeline: read documents into structured elements (with element types like Title, NarrativeText, ListItem, Table and metadata including page numbers), extract plain text, extract tables in HTML/text/JSON formats, split content into overlapping chunks using size-based or semantic (by-title) strategies, and format chunks into arrays ready for embedding APIs (like OpenAI) or vector database records (like LanceDB) with auto-generated IDs and source metadata. PDF parsing supports multiple strategies: auto, fast, hi-res, and OCR-only. No authentication required.

Namespace

Robomotion.DocumentProcessor

Latest version

1.0.16 (17 versions published)

Written in

Python

Categories

Artificial IntelligenceContent & File Management

Document Processor nodes (7)

These are the node IDs an agentic flow or an RPA bot uses, exactly as they appear in the Flow Designer and in flow code. The Toolkit node hands an agent several of the package operations at once.

Platforms

The Document Processor package runs on Windows, macOS and Linux robots. Any Robomotion robot on these operating systems can run a flow that uses it.

How to install the Document Processor package

1. Open Packages in the Robomotion Flow Designer.
2. Search for Document Processor or Robomotion.DocumentProcessor, then click Install on its card.
3. The Document Processor nodes appear in the node palette for everyone in the workspace, because installation is per workspace.

From a terminal, the Robomotion CLI installs the same package:

robomotion install Robomotion.DocumentProcessor

Templates that use Document Processor

One ready-made template uses Document Processor nodes in their flows. Open one to start from a working flow.

• RAG with DeepSeek

  Retrieval-augmented generation over your own documents, with the retrieval handed to the agent instead of done for it. Documents are read, chunked, embedded and stored in LanceDB - an embedded vector database that is just a directory on disk. A DeepSeek Agent then answers questions through a search_knowledge tool built from ordinary flow nodes, writing its own queries and searching as many times as a question needs, with the source document named against every fact.

Related integrations

Packages that share a category or keywords with Document Processor. It is filed under Artificial Intelligence and Content & File Management.

• Document UnderstandingRead documents with your Document Understanding projects: digitize, classify and extract fields and tables, send uncertain documents to human review and export the results.
• Knowledge BaseQuery Robomotion Knowledge Bases from flows. Hybrid (vector + keyword) retrieval over locally cached sqlite-vec indexes, an optional Rerank node (Cohere / Voyage AI / Jina), plus a Knowledge Toolkit for Hermes agents. The same binary doubles as the ingestion worker used by robomotion-agent.
• LanceDBStore and search embeddings in LanceDB - the embedded vector database. Works against a local directory, S3/GCS/Azure object storage or LanceDB Cloud. Vector, full-text and hybrid search, SQL filters, upserts, indexing, plus a Toolkit that hands the whole database to an AI agent.
• PDF ProcessorUse PDF Processor package for streamlined form filling and more.
• PublishingTurn Markdown and HTML into finished documents: EPUB e-books, Word files and print-ready PDFs, plus PDF checks and paperback cover sizes. Works offline.
• Google Document AIProvides a simple interface to access Google Document AI, a service that extracts structured data from documents using machine learning and enables automated document processing.

Document Processor documentation

The Document Processor package reference on docs.robomotion.io documents each node and its properties.