docs-search

You can index local documentation trees into a SQLite FTS5 index and run CLI queries to retrieve relevant snippets with BM25 ranking entirely offline.

Yes, you can index large documentation trees incrementally. The indexer applies SQLite FTS5 with incremental updates to add new docs and maintain search results without full rebuilds.

The best way to find text across thousands of local docs is using a SQLite FTS5 index with BM25 ranking, which provides fast queries and precise highlighting to surface relevant snippets.

No, local documentation search does not require an internet connection. It operates completely offline by indexing files into a local SQLite database and querying them via CLI.

You can check the size and contents of your documentation index by using the info CLI command, which monitors index metrics to help you plan maintenance and manage indexed docs.

BM25 ranking for local documentation retrieval scores search results by term frequency and relevance within the SQLite FTS5 index, ensuring the most pertinent snippets surface first during CLI queries.