1주차 - 동작하는 OS 뼈대#35
Open
wagranungyo wants to merge 8 commits into
Open
Conversation
7 skills, 8 agents, and a paper-os workflow that gauges paper complexity, decides optimal per-stage agent counts, and runs analyzer→detail→code→run→ design→html with feedback gates. Includes LiveEdit example outputs with a verified CPU reproduction demo (ALL PASS). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ROOT now defaults to '.' (session working directory) instead of a hardcoded
absolute Windows path, so the OS runs on any machine / any clone location.
Optionally override via args object { link, root, maxParallel }. README gains
a "다른 PC에서 이어서 작업하기" section.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Move LiveEdit artifacts into output/liveedit_2606.26740/ - Merge 5 feedback_*.md into feedback_summary.md; drop merged detail part files - paper-os.js now writes every run under output/<slug>/ (slug from Triage); intermediate part files are cleaned up after merge - Add output/README.md index and update main README Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Stages and commits working-tree changes with a standardized message (Co-Authored-By trailer, imperative summary). Intended for saving progress and for committing paper-os outputs after a cycle completes. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
paper-os run: JoyAI-VL-Interaction — Real-Time Vision-Language Interaction Intelligence. Complexity=high (detail×3, code×3); all gates PASS (analysis/detail/code/run/html). Feedback consolidated into feedback_summary.md and added to output/README.md index. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Change /code-run default: prepare the ORIGINAL repo for the user to run in their own terminal (copy-paste command block + run guide), instead of Claude auto-executing a reconstructed toy demo - Reconstruction demo demoted to offline-smoke-test fallback only - Rewrite both 05_run.md outputs (joyaivl 9d07596, liveedit cp-cp) as original-repo run guides; logs marked "아직 미실행" - Add CONTEXT.md changelog and record the intent→skill-default principle Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- New /interview skill + agent: before running paper-os, extract hidden assumptions, detect gaps, and present A/B forks, then write the confirmed intent to output/<slug>/00_intent.md (max 2 rounds) - Wire paper-os.js: new Intent phase; every stage reads 00_intent.md and honors it over skill defaults; Triage reuses an existing intent folder - Fix stale code-run stage prompt (was "make toy demo and run it") to match the original-repo / user-runs philosophy - Add joyaivl 00_intent.md from a dry run; log changes in CONTEXT.md Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Add interview stage at the front (skills 7→8, agents 8→9) - Fix code-run description: original-repo, user-runs-in-terminal guide - Reflect per-slug output/<slug>/ layout and 00_intent.md - Add interview→00_intent to the data-flow diagram - Log the doc sync in CONTEXT.md Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
개요
논문 링크 하나로 읽기 → 상세 해설 → 코드 분석 → 실제 실행 → 디자인 → HTML 리포트까지 자동 산출하는 Claude Code 기반 멀티 에이전트 시스템(paper-analysis)입니다.
구성
산출물 예시 (output//)