Judges whether a fix loop's next cycle is still worth spending, from the trend of findings across rounds and the budget used, and reports continue, stop_needs_human or stop_downgrade. Use when Kraft asks it after a fix loop's first cycle, before each further one; it neither reviews nor fixes code.
Install
npx skills add itsOmidKarami/kraft --skill fix-loop-judgenpx skills add itsOmidKarami/kraft --skill fix-loop-judge -a claude-code -gnpx skills add itsOmidKarami/kraft --skill fix-loop-judge -a codex -ggemini skills install https://github.com/itsOmidKarami/kraft.git --path src/kraft/skills/fix-loop-judgegh skill install itsOmidKarami/kraft fix-loop-judgegit clone --depth 1 https://github.com/itsOmidKarami/kraft /tmp/itsOmidKarami__kraft
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/itsOmidKarami__kraft/src/kraft/skills/fix-loop-judge ~/.claude/skills/fix-loop-judge # Claude Code
cp -R /tmp/itsOmidKarami__kraft/src/kraft/skills/fix-loop-judge ~/.agents/skills/fix-loop-judge # Codex, Cursor, Gemini CLI, Copilot, OpenCode