Agent♥︎Age
Catalog

Unstructured Transform

OfficialLive

by Unstructured-IO · TypeScript

Turn documents into structured, AI-ready data by parsing, enriching, chunking, and embedding.

io.unstructured/transform — Model Context Protocol (MCP) Server

This MCP server provides a single integration to transform input files into structured, AI-ready data. It parses documents, enriches content, chunks it, and embeds the results so they can be used directly from an agent session without a separate pipeline.

🛠️ Key Features

  • Parses documents
  • Enriches content
  • Chunks output into partitions
  • Creates embedded representations

🚀 Use Cases

  • Convert PDFs, spreadsheets, scans, and emails into agent-ready data
  • Produce partitioned, enriched, chunked, and embedded output for RAG and vector stores
  • Provide agent memory inputs derived from file content

⚡ Developer Benefits

  • “One call” to obtain output ready for RAG, vector stores, or agent memory
  • Supports “60+ formats” for file ingestion and transformation

⚠️ Limitations

  • Supported formats are stated as “60+,” but specific format details are not included in the provided data