global-projects

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of @liangmi/moi and @liangmi/mo Node.js packages. These are tools developed by the skill author to manage local project structures.
  • [COMMAND_EXECUTION]: The skill utilizes shell commands to interact with the filesystem and external CLI tools. It executes mo list, mo clone, mo fork, and mo-get-root to resolve paths and perform repository operations.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to build shell commands using raw user input (repository names and URLs) without specifying sanitization or validation steps.
  • Ingestion points: User-supplied GitHub repository identifiers (e.g., <owner>/<repo>) and URLs in SKILL.md.
  • Boundary markers: None present to distinguish instructions from untrusted repository data.
  • Capability inventory: Shell command execution via mo and moi CLI tools for repository manipulation in SKILL.md.
  • Sanitization: Absent; the instructions direct the agent to "join the pieces directly" when constructing paths and executing commands, which may allow malicious input to escape intended command arguments.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 12:26 PM
Security Audit — agent-trust-hub — global-projects