flutter-review-request
Pass
Audited by Gen Agent Trust Hub on Feb 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute
git rev-parseto retrieve commit identifiers (SHAs). These commands are read-only and used legitimately to determine the scope of the code changes for review. - [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by interpolating variable content into a subagent prompt.
- Ingestion points: Processes dynamic data from
{WHAT_WAS_IMPLEMENTED},{PLAN_OR_REQUIREMENTS}, and{FILES_CHANGED}variables. - Boundary markers: The prompt uses Markdown headers to separate sections but lacks explicit delimiters or instructions for the subagent to ignore potentially malicious commands embedded within the code or plan descriptions.
- Capability inventory: The skill dispatches tasks to external subagents
flutter-craft:flutter-code-reviewerandflutter-craft:flutter-review-receiveand executes localgitcommands. - Sanitization: No evidence of input validation or sanitization of the implementation details or plan data before they are interpolated into the prompt.
Audit Metadata