Open-source AI personal assistant for planning, writing and personal knowledge, runnable on a Mac or self-hosted with your own model keys.
flagdizero/jenny-android-ai-agent
A local-first personal AI agent for Android with permanent memory, scheduled autonomous tasks, self-written mini-apps, and an optional home-screen launcher.
- Stars
- 109
- Forks
- 26
- Open issues
- 4
- Last commit
- 18 days ago
What is it?
Jenny is an Android app with an embedded Python 3.11 runtime that runs a personal AI agent on the phone itself. It can register as the HOME activity, and a foreground service keeps it running with the screen off so it can send scheduled reminders and proactive messages. Conversations, memory and files are stored in the app's private storage.
The agent builds a Markdown-based memory from recent conversations, and a built-in llm-wiki skill compiles articles, notes and PDFs into cross-linked wiki pages. Users can describe a mini-app in chat and the agent generates its UI, actions and storage, with each action callable as a tool, including from scheduled jobs. It also reads and writes files, executes Python, and searches the web through a hidden Chrome WebView.
You supply your own API key for a hosted provider, or point it at a local model through Ollama or LM Studio to run fully offline. The project is a fork of nanobot and is licensed under AGPL-3.0. It is sideloaded as an APK for Android 8.0 or newer, with no Play Store listing.
Key features
- 01Long-term memory distilled from recent conversations into on-device Markdown files.
- 02Scheduled reminders and recurring jobs that message the user first, even with the screen off.
- 03Chat-driven creation of mini-apps whose actions can be called by the user, the agent, or cron.
- 04Built-in llm-wiki skill that compiles sources into cross-linked Markdown pages with a graph view.
- 05Optional home-screen launcher with a Home gesture and app panel.
- 06Works with 40+ hosted LLM providers via your own key, or local models for offline use.
Use cases
- Keeping a private, persistent assistant that remembers your preferences and context across sessions.
- Building a personal wiki from articles and PDFs you collect.
- Setting up scheduled summaries or reminders that the agent sends to your phone.
- Running an AI assistant fully offline with a local model served by Ollama or LM Studio.
Topics
FAQ
- What is jenny-android-ai-agent?
- A local-first personal AI agent for Android with permanent memory, scheduled autonomous tasks, self-written mini-apps, and an optional home-screen launcher. Jenny is an Android app with an embedded Python 3.11 runtime that runs a personal AI agent on the phone itself. It can register as the HOME activity, and a foreground service keeps it running with the screen off so it can send scheduled reminders and proactive messages. Conversations, memory and files are stored in the app's private storage.
- Is jenny-android-ai-agent free to use? What license does it have?
- jenny-android-ai-agent is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
- Is jenny-android-ai-agent still maintained?
- Actively maintained. The last commit to jenny-android-ai-agent was on Oct 4, 2026.
- What can I use jenny-android-ai-agent for?
- Keeping a private, persistent assistant that remembers your preferences and context across sessions. Building a personal wiki from articles and PDFs you collect. Setting up scheduled summaries or reminders that the agent sends to your phone. Running an AI assistant fully offline with a local model served by Ollama or LM Studio.
- Where can I get jenny-android-ai-agent?
- The source code and installation instructions are on GitHub: https://github.com/flagdizero/jenny-android-ai-agent
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A Rust-based, local-first personal AI agent that runs tasks on your computer and chats through channels like Telegram, Discord, and Slack.
Local-first personal OS agent built on Hermes that logs mood, sleep, meals and habits, then surfaces patterns across them over time.
A local-first personal AI assistant in readable Python, with its own agent loop, SQLite-based memory, and built-in evals.
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.
Self-hosted, local-first personal AI agent with layered memory, tool permissions, workflows and IM channel integration.