This project provides an automated pipeline for researchers to discover, process, and organize academic papers using Claude Code skills and Obsidian. It targets academics and power users overwhelmed by paper volume, offering a significant time-saving benefit by automating literature review tasks and knowledge management.

How It Works

The system employs a multi-stage pipeline powered by Claude Code. It begins by fetching and scoring daily new papers from sources like arXiv, filtering them based on user-defined keywords. A critical review stage uses Claude to summarize and analyze candidate papers, categorizing them for further action. For selected papers, a paper-reader component generates detailed, structured notes, including formulas and figures, and identifies key concepts for a knowledge base. Finally, all outputs are automatically integrated into an Obsidian vault, with a generate-mocs component maintaining linked index pages for easy navigation. This approach automates a complex, time-consuming research workflow, leveraging LLMs for content understanding and generation.

Quick Start & Requirements

• Prerequisites: Claude Code, Obsidian, Python 3.8+, poppler-utils (install via apt or brew), Zotero (optional).
• Installation: Clone the repo, copy skills to ~/.claude/skills/, set the VAULT environment variable to your Obsidian vault path, and create necessary subdirectories within the vault.
• Configuration: Edit ~/.claude/skills/_shared/user-config.json to specify Obsidian vault path, Zotero paths (if used), and keywords for paper filtering and ranking.
• Running: Initiate tasks via natural language commands to Claude, such as "今日论文推荐" for the full pipeline or "读一下这篇论文 [URL/path]" for single papers.
• Links: A video demonstration and Obsidian templates are mentioned but not directly linked.

Highlighted Details

• Automated daily paper discovery, screening, and recommendation generation.
• Full paper parsing, rapid summarization, and critical analysis.
• Concept note generation and linking for knowledge base enrichment.
• Seamless integration with Obsidian for note-taking and knowledge graph maintenance.
• Optional Zotero integration for managing existing literature libraries.

Maintenance & Community

The project encourages community contributions via pull requests and issues. A codex+humanoid branch indicates ongoing adaptation efforts, suggesting a willingness to integrate improvements. No specific community channels (like Discord/Slack) or roadmap links are provided in the README.

Licensing & Compatibility

The project is licensed under the Apache-2.0 license. This license is permissive and generally compatible with commercial use and linking within closed-source projects.

Limitations & Caveats

AI-generated content may contain factual errors or omissions and should be used as an aid rather than a definitive source. Potential bugs and environment-specific adaptation issues may arise. The system is presented as a personal workflow and may require tuning. Default settings do not automatically commit or push changes to a Git-managed Obsidian vault.