Skip to content

Search Modes

sessfind supports four search modes. Use Shift+Tab in the TUI to cycle between them, or pass --method in the CLI.

FTS — Full-Text Search

The default mode. Powered by tantivy with BM25 ranking.

sessfind Full-Text Search mode

shopping                single keyword
shopping assistant      any of these words (OR)
+shopping +assistant    all words required (AND)
"shopping assistant"    exact phrase
shopp*                  prefix wildcard

Tip

FTS is fast and works well for most queries. Use it when you remember specific keywords from a session.

Case-insensitive substring match across content, project name, and title. Good for short, approximate queries.

sessfind Fuzzy search mode

sessfind search "auth" --method fuzzy

Agentic search using your installed AI CLI tools (Claude Code, OpenCode, Copilot). Each detected tool appears as a separate mode in the TUI: LLM (claude), LLM (opencode), etc.

How it works:

  1. You type a natural language query (e.g., "that conversation about fixing CI")
  2. Press Enter to trigger the search
  3. The LLM analyzes your intent and generates optimized FTS queries (synonyms, related terms, multi-language)
  4. sessfind executes each generated query and merges the results

Auto-detection

No extra installation needed. If you have claude, opencode, or copilot on your PATH, the LLM mode appears automatically. The LLM is invoked in headless mode (e.g., claude -p).

Multi-provider

Each detected AI tool becomes its own search mode. You can try the same query with different LLM backends.

ML embedding similarity search. Finds conceptually similar sessions even when exact keywords don't match.

  • Requires the sessfind-semantic plugin
  • Supports Polish and English (and 100+ languages via multilingual-e5-small)
  • Press Enter to trigger (not instant — runs the ML model)
sessfind search "database connection pooling" --method semantic

Warning

Semantic search is slower than FTS or fuzzy because it runs a local ML model. The first run downloads the model (~450MB).