speckit-analyze-zh
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is configured to execute local scripts within the project directory, specifically
.specify/scripts/bash/check-prerequisites.shand.specify/scripts/powershell/check-prerequisites.ps1. These scripts are used to determine the environment configuration and available documentation. - [INDIRECT_PROMPT_INJECTION]: The skill processes multiple external files (
spec.md,plan.md,tasks.md, andconstitution.md) to build a semantic model for analysis. As these files originate from the user's workspace and are processed without defined boundary markers or sanitization, they represent a surface for indirect prompt injection attacks. - [COMMAND_EXECUTION]: The instructions contain specific guidance on escaping user-provided input (
$ARGUMENTS) for shell execution, indicating that input is interpolated into command strings. This poses a potential risk for command injection if the underlying execution platform does not provide robust isolation or sanitization for these arguments.
Audit Metadata