moonbit-refactoring
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the "moon" command-line interface, which is the official toolchain for the MoonBit language. Commands such as "moon check", "moon test", "moon ide find-references", and "moon coverage analyze" are used to perform static analysis, verify code correctness, and generate coverage reports. These are standard operations for development tools provided by the vendor "moonbitlang".
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes external MoonBit source files and markdown documentation. Additionally, it executes local tests via the "moon test" command, which could potentially run malicious code if the project files are compromised.
- Ingestion points: Reads ".mbt" and ".mbt.md" files during the refactoring workflow (SKILL.md).
- Boundary markers: No specific delimiters or warnings to ignore embedded instructions are provided in the skill text.
- Capability inventory: The skill can execute shell commands ("moon"), perform file-system operations, and query IDE metadata (SKILL.md).
- Sanitization: No explicit sanitization or validation of the processed code content is mentioned.
Audit Metadata