Skip to main content

autoacad-prepare

Use when starting a paper project and needing to inventory local code, docs, datasets, constraints, and source materials before literature review or experiment planning.

소스 정보

저장소
zinknovo/autoacad
최근 소스 활동
2026년 9월 28일 12:23
감지된 SKILL.md 언어
영어
스타
0
포크
0

설치 방법

기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.

소스 파일 검토

설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.

SKILL.md 표시 중

SKILL.md
소스 지침 · 읽기 전용 미리보기
name
autoacad-prepare
description
Use when starting a paper project and needing to inventory local code, docs, datasets, constraints, and source materials before literature review or experiment planning.
# AutoAcad Prepare ## Overview Use this stage to establish the paper project's ground truth before any literature or experiment work. ## Workflow 1. Read the project docs first: abstract draft, experiment design notes, whole-paper concept notes, and any constraints file. 2. Inventory code, data, examples, and docs that the paper must describe faithfully. 3. Detect hardware constraints early. Record whether the project is CPU-only, CUDA, or Apple MPS. 4. Identify missing prerequisites: absent docs, missing dataset paths, missing template, or missing baseline implementations. 5. Initialize the canonical project tree with `../scripts/init_paper_project.py` if the project is not structured yet. 6. Write the first stage plan and the initial `PROGRESS.md` entry before moving on. 7. If the idea requires reference codebases, follow `../references/prepare-reference-selection.md` and record the resulting `reference_codebases`, `reference_paths`, and `reference_papers` in `PROGRESS.md`. ## Required Outputs - A concise source-of-truth inventory. - A list of blocked inputs and assumptions. - A confirmed project tree rooted at `<project>-paper/`. - A first-pass environment note in `PROGRESS.md`. - If applicable, `reference_codebases`, `reference_paths`, and `reference_papers` per `../references/prepare-reference-selection.md`.
GitHub에서 보기