Learn

• What is file enrichment?

  The AI-ready file pattern explained — how enrichment differs from OCR, parsing, and RAG.

• What is an LLM-ready file?

  The file format property that lets any AI tool read meaning without re-parsing.

• Enrichment vs. chunking

  Why embedding semantic metadata inside a file is not the same as splitting it for a vector DB.

• OCR once, read forever

  Stop paying to OCR the same file every time a new AI tool touches it.

• Semantic layer for files

  Borrows from BI semantic layers (dbt, Looker) — structured meaning that travels with every file.

• Enrichment vs parsing

  Parsing extracts. Enrichment persists. LLMind lives at the layer above parsers.

• XMP metadata in Python

  Read and write XMP metadata programmatically. Examples using LLMind and Piexif.

• Custom XMP namespace

  Define your own XMP metadata namespace. Portable, file-native, tamper-evident.

• How to give Claude access to files

  Three ways: upload in chat, filesystem MCP, or LLMind-enriched directory. Tradeoffs for each.

• AI agent file access patterns

  4 patterns compared: context dump, RAG stack, MCP-only, and MCP + enriched files.

Explore more

• Use-cases
• Glossary
• Spec