Set task-appropriate output token limits in model requests and handle truncated responses explicitly.
Install
npx skills add alivirgo/Major-AI-Skills --skill max-tokens-ceiling-guardnpx skills add alivirgo/Major-AI-Skills --skill max-tokens-ceiling-guard -a claude-code -gnpx skills add alivirgo/Major-AI-Skills --skill max-tokens-ceiling-guard -a codex -ggemini skills install https://github.com/alivirgo/Major-AI-Skills.git --path skills/max-tokens-ceiling-guardgh skill install alivirgo/Major-AI-Skills max-tokens-ceiling-guardgit clone --depth 1 https://github.com/alivirgo/Major-AI-Skills /tmp/alivirgo__Major-AI-Skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/alivirgo__Major-AI-Skills/skills/max-tokens-ceiling-guard ~/.claude/skills/max-tokens-ceiling-guard # Claude Code
cp -R /tmp/alivirgo__Major-AI-Skills/skills/max-tokens-ceiling-guard ~/.agents/skills/max-tokens-ceiling-guard # Codex, Cursor, Gemini CLI, Copilot, OpenCode