10x-stack-assess

AI agent-friendliness is measured by four quality gates: typed, convention-based, popular in training data, and well-documented. A stack scores against these criteria to determine its suitability for automated agent workflows and identify areas needing compensation strategies.

Assess a brownfield project by running the stack assessment in the project directory to automatically detect stack components. It scores these components against quality gates and generates a comprehensive report detailing gaps and instruction file entries for AI workflows.

Yes, stack evaluation requires recognizable project markers in your current working directory to detect existing components. Providing additional context via foundation PRD files is optional but enhances the accuracy of the quality gate scoring and gap analysis.

Compensation strategies are actionable recommendations generated when stack components do not meet the four quality gates. They provide targeted approaches to bridge gaps in typing, conventions, training data popularity, or documentation for better AI agent performance.

The assessment is limited to detecting stack components from recognizable project markers in the current directory. If a project lacks standard markers or clear conventions, the quality gate scoring and subsequent compensation strategies may be incomplete or less effective.