Score Reality, Order, Completeness, and Human Brief dimensions before coding.
npx skills add https://github.com/gapfDev/workflow --skill me-b4
Please help me install this Agent Skill. Skill: me-b4 Source: https://github.com/gapfDev/workflow/tree/main/c1.2/.claude/skills/me-b4 Command: npx skills add https://github.com/gapfDev/workflow --skill me-b4
ME-B4 provides a concise readiness evaluation before coding to prevent misalignment and rework by explicitly measuring four dimensions: Reality, Order, Completeness, and Human Brief.
Run a self-check that scores Reality, Order, Completeness, and Human Brief to confirm readiness before coding.
High-intent search queries and answers about installing and using this skill.
A pre-code readiness check evaluates project alignment before development starts. It scores dimensions like Reality, Order, Completeness, and Human Brief to prevent misalignment and ensure developers can explain the plan without relying on logs.
Evaluate readiness before coding by applying a structured rubric that scores Reality, Order, Completeness, and Human Brief dimensions. This explicit scoring confirms alignment on constraints and verification requirements, preventing rework.
You need to run a readiness check for a coding task when upfront alignment is critical. It is particularly useful before starting a new feature to validate your understanding and confirm you can explain the implementation plan clearly.
Yes, junior developers can use structured readiness checks for feature development. The four-dimension scoring helps junior-to-senior developers validate their understanding of constraints and alignment before writing code.
The best way to verify project alignment is scoring four dimensions: Reality, Order, Completeness, and Human Brief. This provides traceable decisions and an explicit rubric to confirm readiness for safe, clear work.
A readiness rubric for coding tasks is limited to evaluating upfront alignment and constraints. It does not replace the actual coding process or ongoing quality-assurance testing, serving strictly as a pre-code validation step.
FVossebeldAutomate LLM output evaluation with scoring, pairwise comparisons, and bias mitigation.
tobihagemannAutomate staging, testing, code simplification, AI review, and pull request generation.
richkuoAggregate GitHub milestone issue data into a structured markdown table.
lollipopkitPlan and execute complex features using markdown checklists in zfeats files.
