메뉴
One-shot holistic referee review of the full paper. Issues are written to .review/holistic/ as append-only issue.json files.
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
| name | holistic-review |
| description | One-shot holistic referee review of the full paper. Issues are written to .review/holistic/ as append-only issue.json files. |
| user-invocable | true |
| allowed-tools | Bash, Read, Write, Edit, Glob, Grep, Agent, WebFetch |
Run from the repo root. Requires .review/config.json and .review/parsed/ from Phase 0. Can run before, after, or concurrently with the broad sweep.
Read all parsed sections in order to build a complete picture of the paper. If a PDF is available in the repo, read that too for figures and tables.
Write a structured referee report to .review/holistic_report.md in the voice of a knowledgeable, constructive journal referee. The report should cover:
Summarize what the paper does, its main contribution, and key findings.
Issues that would need to be addressed before publication.
Issues that improve the paper but don't block publication.
Specific questions a referee would ask.
Extract each concern (major and minor) from the holistic report and create issue directories in .review/holistic/:
.review/holistic/
<issue-slug>/
issue.json
Each issue.json:
{
"title": "<issue title>",
"spotter": "holistic",
"source_section": "<most relevant section file>",
"quoted_text": "<most relevant quote from the manuscript>",
"description": "<description from the referee report>",
"holistic_severity": "major|minor",
"created_at": "<ISO timestamp>"
}
Use descriptive slugs for directory names (e.g., single-llm-dependence, missing-robustness-checks).
git add -A
git commit -m "review: holistic review — <N> issues (<M> major, <Mi> minor)"
Report: the full referee report summary, number of issues extracted.
Phase 6 — Generate the REVIEW.md report by walking the append-only review tree. Optionally create GitHub Issues.
Add a custom issue spotter for the paper review. Describe what to look for and this will create the spotter file.
Phase 1 — Run the chunks x issue_spotters cross-product to find candidate issues. Uses edsl for parallel LLM review.
Phase 3 — Deduplicate confirmed issues by writing a dedup.json mapping file. No files are moved or modified.
Draft author responses for each confirmed issue. Writes response.json next to each issue (append-only).
Phase 2 — Investigate candidate issues. Writes investigation.json next to each issue.json with evidence and a confirm/reject verdict.