# AI use cases for Muse Lite (yxq953/MuseLite)

Repo: [yxq953/MuseLite](https://github.com/yxq953/MuseLite) | Use cases: 4 | Page: https://dedupedai.com/use-cases/yxq953-muselite/

MuseLite is an open-source Python-first AI workspace for Android that combines OpenAI-compatible chat, Agent tools, a sandbox, browser tools, scheduled prompts, and optional phone control.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Send an AI prompt at a set time on Android: https://dedupedai.com/use-cases/yxq953-muselite/schedule-prompt-on-android/
- Save Agent summaries to a chosen phone folder: https://dedupedai.com/use-cases/yxq953-muselite/save-agent-results-to-phone-folder/
- Create a reusable Skill from a chat request: https://dedupedai.com/use-cases/yxq953-muselite/reusable-skill-for-meeting-notes/
- Draft and copy chat replies from another app: https://dedupedai.com/use-cases/yxq953-muselite/draft-chat-replies-with-phone-control/

## How to send an AI prompt at a set time on Android with Muse Lite

Repo: [yxq953/MuseLite](https://github.com/yxq953/MuseLite) | Difficulty: Intermediate | Time: 1 hour 30 minutes | Basis: From the README

Build the debug APK, then schedule a prompt that Muse Lite sends in a saved conversation.

**Who it is for:** An Android user who wants an AI prompt sent at a chosen time, such as a daily check-in or reminder draft.

**The problem:** Chat apps wait for you to open them and type, so a prompt you want sent later has to be remembered manually.

**The outcome:** A prompt is sent automatically in the selected conversation at the chosen hour and minute.

### Steps

1. From the repository root on Windows, run `powershell -ExecutionPolicy Bypass -File .\build_windows.ps1`.
2. Install the newest APK from `dist` with the adb command in the README after connecting your device.
3. Open Settings in Muse Lite and enter an OpenAI-compatible HTTPS base URL, a model ID, and an API key.
4. Tap Start New Conversation, type the prompt, tap Schedule beside the input field, choose the hour and minute, then tap Confirm.
5. Leave the app as it is and check that the conversation shows the sent prompt at the selected time.

### Prompt to paste into your agent

```text
In a local clone of the yxq953/MuseLite repository on Windows, run build_windows.ps1 from the repository root to produce a debug APK. Report the full path of the newest file in dist. Verify that the reported file ends in .apk and that the build script exited without errors.
```

### Caveat

The build needs a Windows machine with Python 3.13 and the first run downloads Android SDK components that require accepting licenses.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/yxq953-muselite/schedule-prompt-on-android/

## How to save Agent summaries to a chosen phone folder with Muse Lite

Repo: [yxq953/MuseLite](https://github.com/yxq953/MuseLite) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README

Authorize a phone folder, then have the Agent save and re-read Markdown or other text files there.

**Who it is for:** A user who wants AI-written notes or reports stored in a normal phone folder where other apps can see them.

**The problem:** Files written by the Agent stay in the app's private workspace unless they are exported somewhere visible.

**The outcome:** Markdown summaries appear in a folder you picked, and the Agent can list and read them later.

### Steps

1. Open Settings > 文档保存目录 > 选择手机文件夹, pick a subfolder such as Documents/MuseLite, and confirm Use this folder or Allow.
2. Tap 保存测试文档 to confirm a Markdown file appears in the selected folder.
3. Ask the Agent in chat to save a summary with a file name, for example "把这份总结保存到手机目录，文件名为 总结.md".
4. Ask the Agent to list the folder contents, for example "列出我设置目录里的文件", and check that the new file appears.
5. If access is revoked, select the folder again in Settings; use 断开保存目录 to remove authorization without deleting exported files.

### Prompt to paste into your agent

```text
Using the yxq953/MuseLite app on an Android 11+ device, ask the Agent to save a summary of <source text> as <file name>.md in the authorized document folder. Expected output: the document_save result showing the saved path. Verify that the file appears in the folder you selected in Settings.
```

### Caveat

Android may refuse the storage root, Download root, or Android/data as the folder, so pick a subfolder and expect some providers to show only a folder name.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/yxq953-muselite/save-agent-results-to-phone-folder/

## How to create a reusable Skill from a chat request with Muse Lite

Repo: [yxq953/MuseLite](https://github.com/yxq953/MuseLite) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Use the built-in Skill Creator to turn a repeated task into a skill the Agent can call on demand.

**Who it is for:** A user who repeats the same kind of AI task, such as turning meeting notes into action items, and wants it packaged once.

**The problem:** Re-explaining the same workflow in every chat wastes time and produces uneven results.

**The outcome:** A new enabled skill directory exists in the workspace, and the Agent uses it for matching requests.

### Steps

1. Open Settings > Skill to confirm Skill Creator is enabled; it is installed and enabled on first use.
2. In a conversation, ask: "Create a skill that turns meeting notes into decisions and action items."
3. Wait for the Agent to report the new skill directory and confirm it is enabled.
4. Paste sample meeting notes and invoke the skill explicitly with $skill-name using the name the Agent created.
5. Open Settings > Skill, select the new skill, and review its SKILL.md; fix the YAML name and description if a save is rejected.

### Prompt to paste into your agent

```text
In the yxq953/MuseLite app, ask the Agent to create a skill that turns <task description> into a structured output. Expected output: a skill directory under /var/muselite/workspace/skills with a valid SKILL.md. Verify that the YAML name matches the directory name and that the skill is enabled.
```

### Caveat

Skills only supplement the request and do not grant new tools or permissions, and Linux scripts in a skill run only on Android through the shell sandbox.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/yxq953-muselite/reusable-skill-for-meeting-notes/

## How to draft and copy chat replies from another app with Muse Lite

Repo: [yxq953/MuseLite](https://github.com/yxq953/MuseLite) | Difficulty: Advanced | Time: 1 hour 15 minutes | Basis: From the README

Enable phone control so the Agent reads a visible chat screen and copies a suggested reply.

**Who it is for:** An Android user who gets many chat messages and wants a suggested reply copied to the clipboard for review.

**The problem:** Composing replies in busy chat apps takes effort, and manually copying context into an AI app is slow.

**The outcome:** A suggested reply is copied to the clipboard for you to review and paste yourself.

### Steps

1. Open Settings > Phone Control and enable the Muse Lite accessibility service in Android system settings.
2. Grant notification permission when Android asks, then turn on Allow Agent to operate the phone.
3. Configure a vision-capable model in Settings and turn on image input if you want screenshot help.
4. Enter “读取我正在聊天的界面，推荐一个回复并复制到剪贴板” in the conversation input, then switch to the chat app within the 30 second wait.
5. Paste the copied reply yourself after reviewing it; the Agent does not send messages or paste text.

### Prompt to paste into your agent

```text
In the yxq953/MuseLite app with Phone Control enabled, ask the Agent to read the visible chat in <app name> and copy one suggested reply to the clipboard. Expected output: a short reply reported by the Agent and a clipboard write result. Verify that the clipboard now holds the reply and that no message was sent in the chat app.
```

### Caveat

Phone control only reads visible screen text, so hidden history or apps that do not expose text may need a screenshot with an image-capable model.

Risk, credentials: Accessibility access lets the Agent read visible chat text, and screenshots are sent to your configured model provider, so only use it with a provider and app you trust.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/yxq953-muselite/draft-chat-replies-with-phone-control/