by yxq953
MuseLite is an open-source AI workspace combining chat, Agent tools, browser automation, and scheduled tasks for users who enjoy Muse and Dots.
# Add to your Claude Code skills
git clone https://github.com/yxq953/MuseLiteSee how MuseLite compares with popular alternatives.
MuseLite is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by yxq953. MuseLite is an open-source AI workspace combining chat, Agent tools, browser automation, and scheduled tasks for users who enjoy Muse and Dots. It has 53 GitHub stars.
MuseLite'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/yxq953/MuseLite" and add it to your Claude Code skills directory (see the Installation section above).
MuseLite is primarily written in Python. It is open-source under yxq953 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 MuseLite 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.
MuseLite is a Python-first AI workspace for Android. It combines an OpenAI-compatible chat client, persistent conversations, an execution sandbox, browser tools, scheduled prompts, and optional phone automation in one native Android application.
WSL, Linux, and administrator privileges are not required by build_windows.ps1.
From the repository root, run:
powershell -ExecutionPolicy Bypass -File .\build_windows.ps1
The script creates a local .venv, downloads build tools into cache, stages the Android project, and writes a debug APK to dist. On the first run, accept the Android SDK licenses when prompted.
Install the newest APK on a connected device with:
$apk = Get-ChildItem .\dist\*.apk |
Sort-Object LastWriteTime -Descending |
Select-Object -First 1
& .\cache\tools\android_sdk\platform-tools\adb.exe install -r $apk.FullName
Debug APKs are intended for development and testing. They are not release-signed packages for an app store.
Open Settings after launching MuseLite and enter:
For example, DeepSeek can use https://api.deepseek.com with the model ID deepseek-flash. The provider endpoint is validated before settings are saved. The API key is kept in the app's private storage and encrypted with Android Keystore.
MuseLite returns to the main screen after confirmation. At the selected time, Android opens the saved conversation and sends the prompt that was in the input field when the task was created. An empty draft is not saved as a recent conversation.
Phone control is disabled by default. To enable it:
When enabled, the Agent can inspect the current screen, tap or long-press, type or clear text, swipe, press Back/Home/Recents, open an app by package name, and wait for a target to appear or disappear. Tasks run in a foreground service with a persistent notification and can be stopped from the notification or the chat. The Agent does not regain access after the process is killed.
The first shell_execute call unpacks the bundled Alpine root filesystem. The persistent paths are:
/var/muselite/workspace for user files./var/muselite/memory for saved notes and memory.The minimal image does not include Python or pip. If needed and network access is available, install them inside the sandbox with:
apk add --no-cache python3 py3-pip
The installed packages and files remain in the app's private data directory.
src/muselite_py/ Python UI, Agent, storage, tools, and sandbox
java/com/muselite/python/ Android bridges, services, scheduling, and phone control
template/ Briefcase Android template and native resources
scripts/ Windows build, staging, verification, and asset tools
vendor/ Reproducible Alpine and PRoot inputs
tests/ Python unit tests
Run the Python test suite from the repository root:
$env:PYTEST_DISABLE_PLUGIN_AUTOLOAD = '1'
python -m pytest -q tests
For Android validation, install the generated APK on an Android 11+ arm64 device and verify launch, streaming chat, scheduled prompts, multi-step tool calls, sandbox commands, persistence after restart, browser tools, and phone-control authorization changes.
The bundled PRoot loader, PRoot binary, and Alpine root filesystem under vendor/ are derived from the OpenMinis Android 1.13 release. See LICENSE and THIRD_PARTY_LICENSES.md for project and third-party license information.
MuseLite is released under the GNU General Public License v3.0. See LICENSE for the full text.