lilbee 0.3.8

A local, offline knowledge base. Add documents and code, ask questions grounded in what's actually there.

lilbee

This is an experimental tool and a work in progress. There will be issues with some formats and performance at scale is unknown at this time

A local, offline knowledge base. Add documents and code, ask questions grounded in what's actually there.

---

• Why lilbee
• Demos
• Install
• Quick start
• Interactive chat
• Agent integration
• Supported formats
• Configuration
• How it works

---

Why lilbee

Index your documents and code into a local knowledge base, then ask questions grounded in what's actually there. Most tools like this only handle code. lilbee handles PDFs, Word docs, epics — and code too, with AST-aware chunking.

• Documents and code alike — add anything from a vehicle manual to an entire codebase
• Fully offline — runs on your machine with Ollama and LanceDB, no cloud APIs or Docker
• Works with AI agents — MCP server and JSON CLI so agents can search your knowledge base too

Add files (lilbee add), then ask questions or search. Once indexed, search works without Ollama — agents use their own LLM to reason over the retrieved chunks.

Demos

AI agent using lilbee (opencode)

An AI coding agent shells out to lilbee --json search to ground its answers in your documents.

Interactive local offline chat

[!NOTE] Entirely local on a 2021 M1 Pro with 32 GB RAM.

Model switching via tab completion, then a Q&A grounded in an indexed PDF.

Code index and search

Add a codebase and search with natural language. Tree-sitter provides AST-aware chunking.

JSON output

Structured JSON output for agents and scripts.

Install

Prerequisites

• Python 3.11+
• Ollama — the embedding model (nomic-embed-text) is auto-pulled on first sync if not already installed. If you want to use lilbee as a standalone local chat (no cloud LLM), also pull a chat model:

  ollama pull qwen3:8b     # or llama3, mistral, etc.
  

• Optional (for image OCR): brew install tesseract / apt install tesseract-ocr

First-time download: If you're new to Ollama, expect the first run to take a while — models are large files that need to be downloaded once. For example, qwen3:8b is ~5 GB and the embedding model nomic-embed-text is ~274 MB. After the initial download, models are cached locally and load in seconds. You can check what you have installed with ollama list.

Install

pip install lilbee        # or: uv tool install lilbee

Development (run from source)

git clone https://github.com/tobocop2/lilbee && cd lilbee
uv sync
uv run lilbee

Quick start

# Check version
lilbee --version

# Chat with a local LLM (requires Ollama)
lilbee

# Add documents to your knowledge base
lilbee add ~/Documents/manual.pdf ~/notes/

# Ask questions — answers come from your documents via a local LLM
lilbee ask "What is the recommended oil change interval?"

# Search documents — returns raw chunks, no LLM needed at query time
lilbee search "oil change interval"

# Remove a document from the knowledge base
lilbee remove manual.pdf

# Use a different local chat model (requires ollama pull <model>)
lilbee ask "Explain this" --model qwen3

# Check what's indexed
lilbee status

Interactive chat

Running lilbee or lilbee chat enters an interactive REPL with conversation history, streaming responses, and slash commands:

Command	Description
/status	Show indexed documents and config
/add [path]	Add a file or directory (tab-completes paths)
/model [name]	Show or switch chat model (tab-completes Ollama models)
/version	Show lilbee version
/reset	Delete all documents and data (asks for confirmation)
/help	Show available commands
/quit	Exit chat

Slash commands and paths tab-complete. A spinner shows while waiting for the first token from the LLM.

Agent integration

lilbee can serve as a local retrieval backend for AI coding agents via MCP or JSON CLI. See docs/agent-integration.md for setup and usage.

Supported formats

Format	Extensions	Requires
PDF	.pdf	—
Office	.docx, .xlsx, .pptx	—
eBook	.epub	—
Images (OCR)	.png, .jpg, .jpeg, .tiff, .bmp, .webp	Tesseract
Data	.csv, .tsv	—
Text	.md, .txt, .html, .rst	—
Code	.py, .js, .ts, .go, .rs, .java and 150+ more via tree-sitter (AST-aware chunking)	—

Configuration

All settings are configurable via environment variables:

Variable	Default	Description
LILBEE_DATA	(platform default)	Data directory path
LILBEE_CHAT_MODEL	qwen3:8b	Ollama chat model
LILBEE_EMBEDDING_MODEL	nomic-embed-text	Embedding model
LILBEE_EMBEDDING_DIM	768	Embedding dimensions
LILBEE_CHUNK_SIZE	512	Tokens per chunk
LILBEE_CHUNK_OVERLAP	100	Overlap tokens between chunks
LILBEE_MAX_EMBED_CHARS	2000	Max characters per chunk for embedding
LILBEE_TOP_K	10	Number of retrieval results
LILBEE_SYSTEM_PROMPT	(built-in)	Custom system prompt for RAG answers

CLI also accepts --model / -m, --data-dir / -d, and --version / -V flags.

How it works

Documents are hashed and synced automatically — new files get ingested, modified files re-ingested, deleted files removed. Kreuzberg handles extraction and chunking across all document formats (PDF, Office, images via OCR, etc.), while tree-sitter provides AST-aware chunking for code. Chunks are embedded via Ollama and stored in LanceDB. Ollama uses llama.cpp with native Metal support, which is significantly faster than in-process alternatives like ONNX Runtime — CoreML can't accelerate nomic-embed-text's rotary embeddings, making CPU the only ONNX path on macOS (~170ms/chunk vs near-instant with Ollama's GPU inference). Queries embed the question, find the most relevant chunks by vector similarity, and pass them as context to the LLM.

Data location

Platform	Path
macOS	~/Library/Application Support/lilbee/
Linux	~/.local/share/lilbee/
Windows	%LOCALAPPDATA%/lilbee/

Override with LILBEE_DATA=/path or --data-dir.

License

MIT