The Agent Skills Directory

[SAFE]: No malicious patterns detected. The skill's instructions are focused on providing accurate developer guidance for the MoonBit ecosystem using official tools and documentation.\n- [COMMAND_EXECUTION]: The instructions direct the agent to utilize the official moon CLI for tasks like code validation (moon check), testing (moon test), and API discovery (moon ide doc). These are legitimate development operations.\n- [EXTERNAL_DOWNLOADS]: The skill references trusted official resources such as docs.moonbitlang.com and the package registry mooncakes.io for documentation and dependency management.\n- [PROMPT_INJECTION]: The skill processes user-supplied MoonBit source code and configuration files, creating a surface for indirect prompt injection. This is mitigated by instructions requiring toolchain validation.\n
Ingestion points: User project files, source code (.mbt), and configuration metadata (moon.mod.json, moon.pkg.json).\n
Boundary markers: The agent is instructed to use moon check and moon test for validation and to cite official sources for all claims.\n
Capability inventory: Toolchain execution commands (check, build, test, ide, add, update).\n
Sanitization: Verification of project state and code via the official MoonBit compiler and toolchain.

moonbit-orientation