인터페이스는 현지화되어 있으며, Skill 소개와 설명 문서는 선택한 언어로 AI가 자동 번역합니다. 코드·명령어·경로·전문 용어는 원문을 유지하며 원작자 문서가 우선합니다.
Skills 둘러보기
데이터 및 연구 · SKILL.md

nature-academic-search

nature-academic-search

Search, plan, and organize literature for RL, MARL, LLM, and LLM Agent research. Use for finding papers, building related-work maps, identifying baselines, benchmark provenance, citation gaps, and recent work tracking.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill 문서

GitHub

name: nature-academic-search description: Search, plan, and organize literature for RL, MARL, LLM, and LLM Agent research. Use for finding papers, building related-work maps, identifying baselines, benchmark provenance, citation gaps, and recent work tracking. version: 2.0.0-agent-rl

Academic search skill

Workflow

  1. Translate the user's topic into search axes: problem, method, benchmark, baseline, theory, safety.
  2. Generate keyword sets for RL/MARL/LLM/Agent variants.
  3. Prioritize primary sources: papers, official code, benchmark documentation, venue proceedings.
  4. Build a related-work matrix: paper, contribution, method family, benchmark, limitation, relation to user's work.
  5. Mark which references are foundational, competing, complementary or cautionary.
  6. For recent topics, verify publication dates and versions online when tools are available.

Output format

Search plan:
Keyword groups:
Must-read papers:
Baseline candidates:
Benchmark/dataset sources:
Related-work matrix:
Open citation gaps:

Red lines

Do not fabricate search results. If live search is unavailable, say that the list is based on supplied material or prior knowledge and needs verification.

저장한 Skills