by TimeLovercc
Mojito — the local, evolving alternative to cloud personal AI like Meta Muse.
# Add to your Claude Code skills
git clone https://github.com/TimeLovercc/mojitoSee how mojito compares with popular alternatives.
mojito is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by TimeLovercc. Mojito — the local, evolving alternative to cloud personal AI like Meta Muse. It has 50 GitHub stars.
mojito's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/TimeLovercc/mojito" and add it to your Claude Code skills directory (see the Installation section above).
mojito is primarily written in TypeScript. It is open-source under TimeLovercc on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh mojito against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
Mojito is an evolving app — it rewrites itself for you. Tell it in chat what you want, and a Claude Code agent on your own Mac changes the code and ships the update to your phone, Mac and web app. Out of the box it's a daily planner you host yourself: Today, plans, notes, a feed and a chat.
Experimental / alpha. Mojito is a single-user app, built around one person's day. Most of its code was written by Claude Code sessions working under the rules in
CLAUDE.md,docs/design.mdanddocs/api.md. Expect rough edges, breaking changes and setup steps that assume you're comfortable running your own server.
Today's apps are frozen code: someone else decides what they do, and you wait for the next release. An evolving app keeps only a thin shell of code and hands everything else to an agent that owns the context — your data, your logins, and the app's own source and design. Because the agent knows the app and knows you, it can change the app itself, while you stay in the loop.
Mojito is built on three principles:
| Traditional app | Cloud personal AI (e.g. Meta AI / Muse) | Mojito (evolving app) | |
|---|---|---|---|
| Who decides features | The company that makes it | The company that makes it. The assistant can chat and act, but can't change its own app | You, in chat. The agent changes the app's code |
| How fast it changes | At the next release | At the vendor's next release | Often the same day you ask (see below) |
| Backend | The vendor's servers, with the features hard-coded | The vendor's cloud | A thin hub you host (FastAPI + SQLite), plus an agent on your Mac |
| Where your context lives | Inside each app, separately | In the vendor's cloud: what you connect to it or tell it | On your own machines (your Mac and your hub): your logins, files, mail and calendar, plus the app's own source and design |
| Your data & logins | On the vendor's servers | They go to the vendor's cloud | They stay on machines you control: your Mac and the server you run the hub on. What the agent reads goes to the model provider you choose (today Anthropic) |
| Who it fits | Many people, one design for all | Anyone who wants an assistant without setting anything up | One person who can self-host and wants an app shaped around them. Alpha |
Say it, and it gets done. Some requests change your data, each with Undo. Others change the app itself: the agent writes the code and ships it. Anything that goes out to other people, like an email, is only drafted for you to send.
The honest part: your data stays on machines you control (your Mac and your hub server), and nothing goes to the authors of Mojito. But whatever the agent reads is sent to the model provider you choose for processing (today that's Anthropic, through Claude Code). Risky changes wait for your OK, but only because the maintainer follows the rules in the docs; the code doesn't enforce it yet. More in Security & privacy.
Five requests from one day in the author's own copy, translated and shortened. One sentence in, and this is what showed up: