Skills
Navigate
skills/arenukvern/mcp_flutter/flutter-mcp-toolkit-control/Gen Agent Trust Hub

flutter-mcp-toolkit-control

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by allowing the agent to process untrusted UI data from the target application.
  • Ingestion points: UI data is ingested via semantic_snapshot, wait_for, and hot_reload_and_capture (which includes screenshots).
  • Boundary markers: There are no explicit delimiters or instructions to the agent to ignore or sanitize instructions found within the UI labels or text fields.
  • Capability inventory: The agent can perform sensitive actions such as enter_text, fill_form, and navigate based on the ingested UI data.
  • Sanitization: No sanitization of the semantic tree or visual data is performed.
  • [DYNAMIC_EXECUTION]: The skill facilitates dynamic code execution through Flutter's hot reload mechanism.
  • Evidence: The hot_reload_flutter and hot_restart_flutter tools trigger the Flutter VM to recompile and swap code at runtime. This is an intended feature for development but constitutes dynamic execution of the application's source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 11:35 AM