flutter-mcp

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill references CLI commands for the flutter-mcp-toolkit utility, such as doctor and permissions. These are standard administrative actions for local development tools to verify environment setup and manage required system permissions (e.g., macOS screen recording).
  • [REMOTE_CODE_EXECUTION]: The fmt_evaluate_dart_expression tool is used to read runtime state from the target Flutter application. While this involves dynamic execution, it is the primary intended function for debugging and inspecting a running app.
  • [PROMPT_INJECTION]: The skill facilitates reading external data from the target application (UI tree via fmt_semantic_snapshot and errors via fmt_get_app_errors), creating a surface for indirect prompt injection.
  • Ingestion points: UI snapshots and application error messages from SKILL.md tool references.
  • Boundary markers: None present in the instructions.
  • Capability inventory: UI interaction tools (fmt_tap_widget, etc.), state evaluation, and CLI permission commands.
  • Sanitization: None specified.
  • This surface is considered a functional requirement for the skill's purpose rather than a malicious vector.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 12:42 AM
Security Audit — agent-trust-hub — flutter-mcp