Skills
skills/sgaabdu4/building-flutter-apps/building-flutter-apps/Gen Agent Trust Hub

building-flutter-apps

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses dart analyze to perform architectural audits and linting on the project source code.\n- [COMMAND_EXECUTION]: Evaluation scripts in the tool/ directory execute claude and codex binaries to verify skill behavior and response quality.\n- [EXTERNAL_DOWNLOADS]: The tool/check_upstream_flutter_skills.rb script performs a git clone of the official flutter/skills repository to check for drift from upstream standards. This targets a trusted source.\n- [EXTERNAL_DOWNLOADS]: Instructions and setup guides recommend installing standard Dart and Flutter packages via the pub package manager.\n- [PROMPT_INJECTION]: The skill audits user-provided Dart source code and Markdown files. While no malicious behavior was found, this represents a surface for indirect prompt injection.\n
  • Ingestion points: hooks/scripts/dart_gate.sh and hooks/scripts/preflight_audit.sh (reading .dart and l10n.yaml files).\n
  • Boundary markers: Absent.\n
  • Capability inventory: dart analyze, grep, awk, python3 (subprocess calls in hooks/scripts/preflight_audit.sh).\n
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 06:46 PM
Security Audit — agent-trust-hub — building-flutter-apps