Purpose
This tool provides a desktop graphical interface and command-line utility for syncing and managing AI skills, subagents, and MCP server configurations across multiple development environments like Claude, Cursor, and Codex.

Security Assessment
The overall risk is rated as Medium. The tool does not request dangerous system permissions and no hardcoded secrets were found. However, there are a few code-level concerns. A recursive force deletion command (`rm -rf`) was found inside a test script, which could pose a risk if executed in an unexpected environment or modified maliciously. Additionally, environment variable access occurs within its continuous integration workflows, which is standard for build pipelines but warrants a closer look if you plan to contribute to the project. Users should manually verify exactly what file system paths and directories it interacts with before running it.

Quality Assessment
The project is openly licensed under the MIT license and was actively updated very recently. However, community trust and visibility are currently very low, with only 7 stars on GitHub. The creator explicitly warns that this is a "pre-alpha concept" and a personal project. Users should expect potential bugs, instability, and frequent breaking changes as the tool matures toward a stable release.

Verdict
Use with caution: it is an early-stage, actively updated personal project with low community adoption that contains minor code safety warnings to be aware of.