Search and discover documents in your knowledge base.

Ranks results with two independent rankers, selected explicitly:

  --fuzzy QUERY    Subsequence match on title and key. Tolerant of partial
                   words and dropped characters (auth matches Authentication);
                   it matches a subsequence of characters, not substitutions.
  --lexical QUERY  BM25 full-text match on title and body. Stemmed exact
                   tokens, ranked by term frequency, rarity, and length.

Supplying both fuses the two rankings with Reciprocal Rank Fusion. A bare
positional QUERY still works but is deprecated (defaults to --fuzzy and prints
a warning); use --fuzzy or --lexical instead.

Text queries are bag-of-words: word order is ignored and quoted phrases are
not matched as phrases (find "a b" and find "b a" are identical). --lexical
also drops stop words, so a query of only common words (e.g. "the") matches
nothing.

Without a query, documents are sorted by popularity (incoming reference and
inclusion count); with --sort, frontmatter sort wins and the query acts as a
membership filter.

Filter shorthand flags map onto the query operators (--key, --included-by,
--references, --referenced-by, etc.). Use --filter for inline YAML predicates
such as --filter 'status: draft'.
