Direct answer
Metadata is structured information about an artifact. A webpage can use meta elements; images can contain EXIF, IPTC, or XMP; PDFs and office files can store titles, authors, subjects, keywords, and custom properties.
Why metadata can be overlooked
People often review the primary rendered content while software extracts metadata automatically. This creates a second information plane that may be useful for indexing, accessibility, provenance, or puzzles.
Machine processing
Different parsers expose different metadata fields. A safe processing pipeline should keep metadata provenance separate from visible body text rather than silently concatenating everything into one instruction-like stream.
Puzzle use
FugitiveAI uses harmless metadata-like clues to teach where information can reside and how source-aware inspection differs from visual inspection.
Frequently asked questions
Is metadata visible to normal users?
Some metadata is surfaced by applications, but much of it is not part of the primary rendered content.
Can metadata contain text?
Yes. Many formats support free-text fields such as title, subject, description, comment, or keywords.