self-review — Claude Code skill: review and install
This skill reviews an academic paper with the NeurIPS form, three reviewer roles, and score aggregation. Use it to assess a draft before submission.
npx skills add lingzhi227/agent-research-skills --skill self-review -a claude-codeWhat self-review does
The skill takes a PDF or .tex path. For a PDF, it extracts the text and runs three separate reviews: a fair but strict reviewer, a highly critical reviewer, and an open-minded reviewer. Each review follows the NeurIPS form. The skill then combines the results, averages the scores, applies weighted scoring, and writes a report with shared strengths, weaknesses, author questions, and specific revision points.
The package includes extract_pdf_text.py and parse_pdf_sections.py. The first script can use pymupdf4llm, pymupdf, or pypdf; the second identifies the title and section text in a PDF. Section parsing requires pymupdf, while text extraction uses the listed Python packages. The workflow also checks for Abstract, Introduction, Methods/Approach, Experiments/Results, and Discussion/Conclusion. Missing sections lower the review scores.
What it helps with
- review a paper before NeurIPS submission
- extract text from a PDF for analysis
- compare three independent reviewer scores
- check whether required paper sections are present
- turn review findings into concrete revisions
Example prompts for the agent
- Review
paper.pdfwith the NeurIPS form and produce a consolidated report. - Assess
paper.texthrough three reviewer perspectives, then compare their scores. - Check
paper.pdffor required sections and list the most specific changes to make.
How to install self-review
1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill self-review -a claude-code2Manually: put the files in place
~/.claude/skills/self-review/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/self-review/
1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill self-review -a codex2Manually: put the files in place
.agents/skills/self-review/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill self-review -a cursor2Manually: put the files in place
.cursor/skills/self-review/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lingzhi227/agent-research-skills self-review2Manually: put the files in place
.github/skills/self-review/SKILL.md1Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill self-review -a gemini-cli2Platform CLI
gemini skills install https://github.com/lingzhi227/agent-research-skills --path skills/self-reviewManually: put the files in place
.devin/skills/self-review/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lingzhi227/agent-research-skills --skill self-review -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about self-review
How do I install self-review in Claude Code?
Run npx skills add lingzhi227/agent-research-skills --skill self-review -a claude-code. The installed skill file is at ~/.claude/skills/self-review/SKILL.md.
What does self-review do?
It evaluates a PDF or .tex paper with the NeurIPS review form, runs three reviewer personas, combines their findings, and produces scores plus revision feedback.
What is required to parse a PDF?
The skill includes extract_pdf_text.py and parse_pdf_sections.py. PDF section parsing requires pymupdf; text extraction can use pymupdf4llm, pymupdf, or pypdf.
