paper-reviewer-pro

What This Skill Does

Paper Review Pro is a sophisticated AI-driven research assistant designed to accelerate the literature review process for scientists and students. It performs multi-source retrieval from arXiv and Semantic Scholar, intelligently filters results based on relevance and authority, and provides structured summaries. By integrating the CCF (China Computer Federation) rating system, it helps researchers identify high-impact publications while automatically generating professional BibTeX files for seamless citation management in tools like Zotero or Mendeley.

Installation

To integrate Paper Review Pro into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/alfredliang11/paper-review-pro Ensure your local environment is configured by running python scripts/config.py within the skill directory to set up necessary API keys and database paths.

Use Cases

• Systematic Literature Reviews: Quickly build a curated library of the most influential papers in a specific sub-field without manual searching.
• Trend Tracking: Identify the latest developments in 2024-2025 research by filtering for the most recent preprints and published articles.
• Academic Writing: Automatically generate structured summaries (problem, method, results, innovation) to synthesize complex research into accessible notes.
• Cross-Domain Discovery: Utilize the semantic expansion feature to find adjacent research areas or alternative methodologies that might be overlooked in a standard keyword search.

Example Prompts

1. "Perform a literature review on 'Large Language Model reasoning' for papers published in 2024, keep the top 5 most authoritative, and export the BibTeX."
2. "Search for the latest research on 'Graph Neural Networks' and give me a structured summary of the top 3 papers focusing on their specific innovations."
3. "Retrieve 10 recent papers related to 'Diffusion Models' and generate a research report including CCF A-class publications."

Tips & Limitations

• Optimization: For the best results, use precise keywords. If the initial result set is too broad, increase the keep_topk parameter to narrow down to the absolute core literature.
• CCF Reliance: Note that the CCF ranking specifically targets computer science disciplines; results in interdisciplinary fields may show different authority indicators.
• Authority Weights: The authority score defaults to 0.3; if you only want the most prestigious papers, consider adjusting the weighting configurations in the system settings to prioritize CCF-A publications.
• Reliability: Always verify the generated BibTeX content against the original publisher source before final submission in high-stakes academic documents.