skills/mizchi/skills/moonbit-practice/Gen Agent Trust Hub

moonbit-practice

Pass

Audited by Gen Agent Trust Hub on Apr 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the official MoonBit CLI installation script from cli.moonbitlang.com in its CI templates.
  • [REMOTE_CODE_EXECUTION]: The toolchain installation instructions involve fetching and executing a remote shell script via bash.
  • [COMMAND_EXECUTION]: The skill provides examples for running MoonBit toolchain commands, nix shell commands, and various build-time scripts.
  • [DATA_EXFILTRATION]: The guide mentions libraries such as @x/fs and @x/sys which provide access to the file system and environment variables.
  • [PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection within its CI workflow templates.
  • Ingestion points: CI configuration files in the assets/ directory and SKILL.md.
  • Boundary markers: No explicit markers are used in the templates to delimit external data.
  • Capability inventory: Includes subprocess execution (via moon and pnpm) and file system operations.
  • Sanitization: The provided examples do not include input sanitization for data processed in CI pipelines.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 27, 2026, 03:45 AM