DiscordSign up
SKILLS.SH

self-review

Automatically review an academic paper using the NeurIPS review form with three reviewer personas, ensemble scoring, and reflection refinement. Extracts text…

INSTALLATION
npx skills add https://github.com/lingzhi227/agent-research-skills --skill self-review
Run in your project or agent environment. Adjust flags if your CLI version differs.

SKILL.md

Self-Review

Review an academic paper using a structured review form with multiple reviewer personas.

Input

  • $ARGUMENTS — Path to PDF file or .tex file

Scripts

Extract text from PDF

python ~/.claude/skills/self-review/scripts/extract_pdf_text.py paper.pdf --output paper_text.txt

python ~/.claude/skills/self-review/scripts/extract_pdf_text.py paper.pdf --format markdown

Tries pymupdf4llm (best) → pymupdf → pypdf. Install: pip install pymupdf4llm pymupdf pypdf

Parse PDF into structured sections

python ~/.claude/skills/self-review/scripts/parse_pdf_sections.py \

  --pdf paper.pdf --output sections.json

Extracts title (via font size), section headings, and section text. Requires: pip install pymupdf

Key flags: --format text, --verbose

Workflow

Step 1: Load Paper

  • If PDF: use extract_pdf_text.py to extract text
  • If .tex: read the LaTeX source directly

Step 2: Three-Persona Review

Run three independent reviews using different personas (from references/review-form.md):

  • Harsh but fair reviewer: Expects good experiments that lead to insights
  • Harsh and critical reviewer: Looking for impactful ideas in the field
  • Open-minded reviewer: Looking for novel ideas not proposed before

For each persona, generate a review following the NeurIPS review JSON format in references/review-form.md.

Step 3: Reflection Refinement (up to 3 rounds per reviewer)

After each review, apply the reflection prompt: re-evaluate accuracy and soundness, refine if needed. Stop when "I am done".

Step 4: Aggregate

  • Combine all three reviews
  • Average numerical scores (round to nearest integer)
  • Synthesize a meta-review finding consensus
  • Weight scores using AgentLaboratory weights: Overall (1.0), Contribution (0.4), Presentation (0.2), others (0.1 each)

Step 5: Actionable Report

Output format:

## Review Summary

- **Overall Score**: X/10 (Weighted: Y/10)

- **Decision**: Accept / Reject

- **Confidence**: Z/5

## Strengths (consensus across reviewers)

1. ...

2. ...

## Weaknesses (consensus across reviewers)

1. ...

2. ...

## Questions for Authors

1. ...

## Specific Suggestions for Improvement

1. [Section X, Page Y]: ...

2. [Section Z, Page W]: ...

## Score Breakdown

| Dimension | R1 | R2 | R3 | Avg |

|-----------|----|----|-----|-----|

| Overall | ... | ... | ... | ... |

| Contribution | ... | ... | ... | ... |

| ... | ... | ... | ... | ... |

References

  • NeurIPS review form, scoring weights, personas, reflection prompts: ~/.claude/skills/self-review/references/review-form.md
  • PDF text extraction: ~/.claude/skills/self-review/scripts/extract_pdf_text.py

Missing Sections Check

You MUST verify that all required sections are present: Abstract, Introduction, Methods/Approach, Experiments/Results, Discussion/Conclusion. Reduce scores if any are missing.

Related Skills

BrowserAct

Let your agent run on any real-world website

Bypass CAPTCHA & anti-bot for free. Start local, scale to cloud.

Explore BrowserAct Skills →

Related skills

launch-strategy
3/3

Structured framework for planning product launches, feature announcements, and go-to-market strategies. Organizes launches across five phases (internal, alpha, beta, early access, full) with specific actions and goals for each stage Uses the ORB framework to balance owned channels (email, blog, community), rented channels (social, app stores), and borrowed channels (guest posts, influencers, partnerships) Includes dedicated Product Hunt strategy with pre-launch preparation, launch-day tactics, and post-launch conversion guidance Provides launch checklist covering pre-launch setup, launch-day coordination, and post-launch user education and retention Emphasizes treating launches as ongoing moments rather than one-time events, with guidance on spacing feature announcements and maintaining momentum

coreyhaines31
SKILLS.SH
56.2K29.8K
2026-05-21
competitor-alternatives
2/3

SEO-optimized competitor comparison and alternative pages that position your product against rivals. Covers four page formats: singular alternatives, plural alternatives, you vs. competitor, and competitor vs. competitor comparisons Includes centralized competitor data architecture for consistent, maintainable information across all comparison pages Provides structured research process covering product features, pricing, reviews, and customer feedback with quarterly update cadence Emphasizes honest positioning that acknowledges competitor strengths and clearly defines who each product serves best

coreyhaines31
SKILLS.SH
54.3K29.8K
2026-05-21
firecrawl-search
1/3

Web search with optional full-page content extraction from results. Returns real search results as JSON with optional --scrape flag to fetch complete page markdown for each result, avoiding redundant fetches Supports filtering by source type (web, images, news), category (GitHub, research, PDF), time range (past hour/day/week/month/year), location, and country Use --limit to control result count and --scrape-formats to customize output formats when extracting full content Part of a workflow escalation pattern: search first to discover URLs, then use dedicated scrape/map/crawl skills for deeper extraction

firecrawl
SKILLS.SH
43.3K414
2026-05-21
firecrawl-crawl
2/3

Bulk extract content from entire websites or site sections with depth and path filtering. Crawls pages following links up to configurable depth limits and page counts, with path inclusion/exclusion filters to scope extraction Supports async job polling or synchronous waiting with progress display via --wait and --progress flags Offers concurrency control, request delays, and JSON output formatting for integration into agent workflows Part of a four-step escalation pattern: search → scrape → map → crawl, used when single-page extraction is insufficient

firecrawl
SKILLS.SH
42.6K414
2026-05-21

Stop writing automation&scrapers

Install the CLI. Run your first Skill in 30 seconds. Scale when you're ready.

Start free
free · no credit card