flutter-expert

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): No malicious patterns or security risks were detected within the skill instructions or metadata.
  • [PROMPT_INJECTION] (SAFE): The instructions follow standard persona-defining patterns and do not contain attempts to bypass agent safety filters or override core constraints.
  • [DATA_EXFILTRATION] (SAFE): No access to sensitive local files or hardcoded credentials was identified.
  • [REMOTE_CODE_EXECUTION] (SAFE): The skill does not include commands for downloading external scripts or executing dynamic code.
  • [INDIRECT_PROMPT_INJECTION] (SAFE): While the skill is designed to process user input (Ingestion), it lacks any dangerous capabilities such as file-writing, network operations, or subprocess execution, rendering the attack surface non-exploitable.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:01 PM