inode
FeaturedA privacy-focused CLI for storing and retrieving notes, secrets, and commands through natural-language semantic search.
Overview
inode is a command-line knowledge base you talk to in plain English. Instead of remembering exact filenames or grepping through scattered notes, you ask for what you need — “the staging database password” or “how I deployed the worker last time” — and inode finds it by meaning rather than exact keywords.
It is local-first by design: everything runs on your machine with no API keys or internet required, so your notes and secrets never leave your laptop. When you want higher-quality results, you can opt into cloud backends without changing how you use it.
At a glance
- Language
- Go
- Default storage
- SQLite + sqlite-vec
- Optional backends
- PostgreSQL/pgvector · Claude · Voyage AI
- Security
- AES-256-GCM, local-first
- Integrations
- MCP (Claude Code, Cursor)
- Platforms
- macOS · Linux · Windows
How it works
01Semantic search
Built in Go with vector embeddings and LLM inference for natural-language retrieval. Content is auto-classified into nine strict categories — credentials, commands, snippets, runbooks, and more — so what you store stays organized and what you ask for comes back precise.
02Local-first architecture
Uses SQLite + sqlite-vec by default, with no API keys or internet required. Optional PostgreSQL/pgvector, Claude API, and Voyage AI backends are available when you want higher-quality embeddings and results.
03Security by default
Sensitive values are encrypted at rest with AES-256-GCM and masked in output. Ollama provides zero-cost local embeddings and inference, so nothing sensitive is sent anywhere unless you explicitly opt in.
04AI integration
Ships a read-only Model Context Protocol (MCP) server so tools like Claude Code and Cursor can query your knowledge base directly. Cross-platform binaries are available for macOS, Linux, and Windows.
Want the code?
inode is open source and built in public.