A local-first personal AI assistant in readable Python, with its own agent loop, SQLite-based memory, and built-in evals.
oratis/LISA
Open-source AI personal assistant for planning, writing and personal knowledge, runnable on a Mac or self-hosted with your own model keys.
- Stars
- 175
- Forks
- 2
- Open issues
- 9
- Last commit
- 4 days ago
What is it?
LISA is a personal AI assistant that helps with planning a day, drafting messages, and keeping a personal knowledge base. It can be used through a signed Mac app, a CLI with a self-hosted web workspace, or the hosted LISA Cloud service. Self-hosted setups support Anthropic, OpenAI, Gemini, OpenAI-compatible services and local models such as Ollama and LM Studio.
Beyond chat, LISA keeps file-backed identity, memories and journals within an instance, and can save links, ingest articles and video transcripts, and search a linked wiki. It can also observe supported coding agent sessions such as Claude Code, Codex, OpenCode and Aider. Optional read-only mail, MCP tools, and messaging channels including Telegram, Discord, Slack, Feishu and iMessage require explicit setup.
Autonomous features such as goals, routines and watchers are opt-in, and actions without an approval integration are denied. Local data is stored under ~/.lisa by default, while choosing a remote model sends relevant context to that provider.
Key features
- 01Chat with streaming replies, tool activity, history and retry.
- 02Persistent identity, memories and journals within a single instance.
- 03Save links, ingest articles and video transcripts, and search a linked wiki.
- 04Observe and coordinate Claude Code, Codex, OpenCode and Aider sessions.
- 05Supports Anthropic, OpenAI, Gemini, OpenAI-compatible APIs and local models.
- 06Opt-in Warden approval layer with scoped grants and audit records for web chat.
Use cases
- Planning a day and drafting messages or writing from a personal assistant.
- Building a searchable personal knowledge base from saved links and transcripts.
- Monitoring and dispatching work across local coding agent sessions on a Mac.
- Running a self-hosted assistant with a local model such as Ollama.
Quick start
- 1
Install the CLI with Homebrew
brew install oratis/tap/lisa - 2
Or install the CLI with npm
npm install -g @oratis/lisa - 3
Add your model API key to ~/.lisa/config.env (create the folder first)
mkdir -p ~/.lisa - 4
Start the web workspace at http://localhost:5757
lisa serve --web
Things to know
- Requires Node 22.19+ or a provider API key, or a local model, for the CLI and self-hosted setup.
- Hosted LISA Cloud and your Mac instance keep separate histories, memories and credentials.
- The iOS app is not yet publicly approved on the App Store.
- Local tools only work while your Mac is running and reachable.
Topics
FAQ
- What is LISA?
- Open-source AI personal assistant for planning, writing and personal knowledge, runnable on a Mac or self-hosted with your own model keys. LISA is a personal AI assistant that helps with planning a day, drafting messages, and keeping a personal knowledge base. It can be used through a signed Mac app, a CLI with a self-hosted web workspace, or the hosted LISA Cloud service. Self-hosted setups support Anthropic, OpenAI, Gemini, OpenAI-compatible services and local models such as Ollama and LM Studio.
- Is LISA free to use? What license does it have?
- LISA is open source under the MIT license. Check the license terms before using it in commercial products.
- Is LISA still maintained?
- Actively maintained. The last commit to LISA was on Oct 19, 2026.
- What can I use LISA for?
- Planning a day and drafting messages or writing from a personal assistant. Building a searchable personal knowledge base from saved links and transcripts. Monitoring and dispatching work across local coding agent sessions on a Mac. Running a self-hosted assistant with a local model such as Ollama.
- How do I install LISA?
- Install the CLI with Homebrew: brew install oratis/tap/lisa Or install the CLI with npm: npm install -g @oratis/lisa Add your model API key to ~/.lisa/config.env (create the folder first): mkdir -p ~/.lisa Start the web workspace at http://localhost:5757: lisa serve --web
- Where can I get LISA?
- The source code and installation instructions are on GitHub: https://github.com/oratis/LISA
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A self-hosted personal AI assistant that runs on your machine or in the cloud and connects to chat apps like DingTalk, Lark, Discord and Telegram.
Self-hosted AI personal assistant that chats via Telegram, runs scheduled tasks, and connects to MCP servers and host tools.
Self-hosted, local-first AI companion runtime with a Vietnamese-first character, local SQLite memory, Telegram/Zalo channels, skills, and permission-gated tools.
A lightweight self-hosted AI agent written in Go that runs on your server and works through Telegram, using Anthropic, OpenAI or OpenRouter models.
A lightweight, self-hosted personal agent app with a TUI and GUI, local workspace tools, human approval, persistent sessions, memory, and Skill/Plugin/MCP extensions.