← Back to home

Ask Your Archive

Query your archive using natural language, powered by local LLMs via Ollama.

Requirements

  • Ollama installed and running locally
  • A model pulled (e.g., ollama pull llama3)

Usage

# Basic query
archivist /path/to/vault --ask "What did I discuss about Python?"

# Specify model
archivist /path/to/vault --ask "Summarize my React conversations" \
  --ask-model llama3

# Stream response
archivist /path/to/vault --ask "Find my API key discussions" \
  --ask-stream

Security

Localhost only: Archivist only connects to Ollama on 127.0.0.1. External hosts are rejected at runtime. This is enforced at the CLI layer before any network connection is attempted.

Options

FlagDescriptionDefault
--ask "question"Your natural language query--
--ask-modelOllama model to usellama3
--ask-streamStream tokens as they generatefalse
--ollama-portCustom Ollama port11434