Skip to main content

Key takeaways

  • Unstructured Transform enables you to automate Unstructured with plain language instead of code or a custom web interface.
  • The Unstructured agent guide gives agents key context they need to automate Unstructured.
  • This documentation website provides:
    • Industry-standard llms.txt and llms-full.txt files, for rapid agent information retrieval and LLM training.
    • Files that support emerging standards for agent skill discovery and agent-to-agent communication: agent skills discovery, agent skills indexing, individual agent skills, and agent-to-agent (A2A) cards.
    • Markdown (.md) versions of all pages, for faster information consumption and usage by agents and LLMs. For example, view this page in .md format.
    • Continuous editorial enhancements to improve emerging generative engine optimization (GEO) guidance for better agent information retrieval and citation, such as:
      • Adding more “key takeaways” sections.
      • Presenting more headings as reader questions.
      • Reducing word counts and complexity.
      • Streamlining user procedures.
      • Moving non-critical information out of the main flow of information.
      • Adding more information designed especially to help train and guide agents.

Next steps

Unstructured Transform quickstart with Claude

Use Unstructured with Claude and plain language prompts to quickly go from input file to structured data.

Unstructured Transform overview

How Transform works and how to use it.

Agent guide

Give your agents critical information to gain immediate knowledge and context about how to automate Unstructured.