메뉴
Repository guide for the LMTI multi-skill package. Use when working inside this repo to find the installable skills, human documentation, and scoring implementation quickly.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Self-assess the current LLM with the LMTI-60 questionnaire, answer all 60 questions without user interaction, and return the full answer sheet in strict JSON for later scoring.
Score an LMTI-60 selftest result, compute dimension lean, dominant and secondary types, detect inconsistency from cross-reference clusters, and render the output as markdown, JSON, or HTML.
| name | lmti-repo |
| description | Repository guide for the LMTI multi-skill package. Use when working inside this repo to find the installable skills, human documentation, and scoring implementation quickly. |
This repository contains two installable skills, not one:
.agents/skills/lmti-selftest.agents/skills/lmti-scorelmti-selftest when the task is to answer the LMTI-60 questionnaire and emit strict JSON.lmti-score when the task is to score a prior LMTI answer sheet.README.md, INSTALL.md, docs/.agents/skills/lmti-selftest/references/questionnaire.json.agents/skills/lmti-score/references/framework.json.agents/skills/lmti-score/scripts/score_lmti.pyKeep human documentation at repo root or in docs/.
Keep installable skill folders lean:
SKILL.mdagents/openai.yamlreferences/scripts/ only when determinism is required