flutter-init

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface where user-provided input is used to construct shell commands, creating a command injection risk.
  • Ingestion points: User-provided values for '폴더명' (folder name), '프로젝트명' (project name), '조직명' (organization name), and '엔티티명' (entity name) in SKILL.md Steps 1 and 2.
  • Boundary markers: None provided to delimit user input from shell command structures.
  • Capability inventory: Execution of flutter create, flutter pub get, and dart run build_runner via the agent's shell capabilities.
  • Sanitization: No specific sanitization or validation logic is defined to prevent the use of shell control characters (e.g., ;, &&, |) in the input fields.
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to execute multiple shell commands (flutter create, dart run build_runner, flutter analyze) to set up the development environment. While these are standard developer operations, the execution flow depends on non-validated user inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 12:52 PM
Security Audit — agent-trust-hub — flutter-init