Skip to main content
KiwiFS provides multiple search strategies. You can use full-text search for keyword matching, vector search for semantic similarity, or metadata queries for structured filtering. BM25-ranked full-text search powered by SQLite FTS5.
string
required
Search query. Supports boolean operators and phrase matching.
number
Similarity threshold (0–1) for “did you mean?” suggestions when the query yields few results.

Query syntax

Search results weighted by trust signals in frontmatter. Pages marked as verified or source-of-truth rank higher.
string
required
Search query (same syntax as full-text search).
Add status: verified or source-of-truth: true to your page frontmatter to boost that page in trust-ranked results.
Vector similarity search using embeddings. Requires a vector search provider configured in .kiwi/config.toml.
string
required
Natural language query.
integer
default:"10"
Maximum number of results to return.
Semantic search requires vector embeddings. Configure an embedder provider (OpenAI, Ollama, Cohere, etc.) in your .kiwi/config.toml under [search.vector].
Fuse FTS and semantic ranks with Reciprocal Rank Fusion. Degrades to lexical-only when no vector index exists.
string
Set to hybrid to enable RRF fusion.
string
required
Search query.
integer
default:"10"
Maximum number of results.
integer
default:"60"
RRF constant. Higher values flatten rank differences.

Context pack

Retrieve pages and pack extracted sections into a token budget. The pack never summarizes. Overflow is listed so you can see what was dropped.
string
required
Retrieval query.
integer
Maximum tokens in the packed context.
integer
Cap on candidate pages before packing.
string
Restrict retrieval to a directory.

Metadata query

Query pages by their frontmatter fields using JSON path syntax.
string
required
JSON path filter expression (e.g. $.status=draft).
string
JSON path to the field to sort by (e.g. $.updated).
string
default:"asc"
Sort order: asc or desc.
Combine metadata queries with the DQL endpoint for more complex structured queries across your knowledge base.
Last modified on August 20, 2026