claude-scholar

<div align="center"> <img src="LOGO.png" alt="Claude Scholar Logo" width="100%"/> <p> <a href="https://github.com/Galaxy-Dawn/claude-scholar/stargazers"><img src="https://img.shields.io/github/stars/Galaxy-Dawn/claude-scholar?style=flat-square&color=yellow" alt="Stars"/></a> <a href="https://github.com/Galaxy-Dawn/claude-scholar/network/members"><img src="https://img.shields.io/github/forks/Galaxy-Dawn/claude-scholar?style=flat-square" alt="Forks"/></a> <img src="https://img.shields.io/github/last-commit/Galaxy-Dawn/claude-scholar?style=flat-square" alt="Last Commit"/> <img src="https://img.shields.io/badge/License-MIT-green?style=flat-square" alt="License"/> <img src="https://img.shields.io/badge/Claude_Code-Compatible-blueviolet?style=flat-square" alt="Claude Code"/> </p> <strong>Language</strong>: <a href="README.md">English</a> | <a href="README.zh-CN.md">中文</a> </div> > Personal Claude Code / **OpenCode** configuration for academic research and software development — covering the full research lifecycle from ideation to publication. ## News - **2026-02-25**: **Codex CLI** support — added `codex` branch supporting [OpenAI Codex CLI](https://github.com/openai/codex) with config.toml, 40 skills, 14 agents, and sandbox security - **2026-02-23**: Added `setup.sh` installer — safe merge into existing `~/.claude`, auto-backup `settings.json`, smart hooks/mcpServers/plugins merge - **2026-02-21**: **OpenCode** support — Claude Scholar now supports [OpenCode](https://github.com/opencode-ai/opencode) as an alternative CLI; switch to the `opencode` branch for OpenCode-compatible configuration <details> <summary>View older changelog</summary> - **2026-02-20**: Bilingual config — translated `CLAUDE.md` to English for international readability; added `CLAUDE.zh-CN.md` as Chinese backup; Chinese users can switch with `cp CLAUDE.zh-CN.md CLAUDE.md` - **2026-02-15**: Zotero MCP integration — added `/zotero-review` and `/zotero-notes` commands, updated `research-ideation` skill with Zotero integration guide, enhanced `literature-reviewer` agent with Zotero MCP support for automated paper import, collection management, full-text reading, and citation export - **2026-02-14**: Hooks optimization — restructured `security-guard` to two-tier system (Block + Confirm), `skill-forced-eval` now groups skills into 6 categories with silent scan mode, `session-start` limits display to top 5, `session-summary` adds 30-day log auto-cleanup, `stop-summary` shows separate added/modified/deleted counts; removed deprecated shell scripts (lib/common.sh, lib/platform.sh) - **2026-02-11**: Major update — added 10 new skills (research-ideation, results-analysis, citation-verification, review-response, paper-self-review, post-acceptance, daily-coding, frontend-design, ui-ux-pro-max, web-design-reviewer), 7 new agents, 8 research workflow commands, 2 new rules (security, experiment-reproducibility); restructured CLAUDE.md; 89 files changed - **2026-01-26**: Rewrote all Hooks to cross-platform Node.js; completely rewrote README; expanded ML paper writing knowledge base; merged PR #1 (cross-platform support) - **2026-01-25**: Project open-sourced, v1.0.0 released with 25 skills (architecture-design, bug-detective, git-workflow, kaggle-learner, scientific-writing, etc.), 2 agents (paper-miner, kaggle-miner), 30+ commands (including SuperClaude suite), 5 Shell Hooks, and 2 rules (coding-style, agents) </details> ## Introduction Claude Scholar is a personal configuration system for Claude Code CLI, providing rich skills, commands, agents, and hooks optimized for: - **Academic Research** - Complete research lifecycle: idea generation → experimentation → results analysis → paper writing → review response → conference preparation - **Software Development** - Git workflows, code review, test-driven development, ML project architecture - **Plugin Development** - Skill, Command, Agent, Hook development guides with quality assessment - **Project Management** - Planning documents, code standards, automated workflows with cross-platform hooks ## Quick Navigation | Topic | Description | |-------|-------------| | 🚀 [Quick Start](#quick-start) | Get up and running in minutes | | 📚 [Core Workflows](#core-workflows) | Paper writing, code organization, skill evolution | | 🛠️ [What's Included](#whats-included) | Skills, commands, agents overview | | 📖 [Installation Guide](#installation-options) | Full, minimal, or selective setup | | 📦 [MCP Setup](#mcp-server-setup-optional) | Zotero MCP for research workflows | | 🔧 [Project Rules](#project-rules) | Coding style and agent orchestration | ## Core Workflows ### Primary Workflows Complete academic research lifecycle - 7 stages from idea to publication. #### 1. Research Ideation (Zotero-Integrated) End-to-end research startup from idea generation to literature management: **Tools**: `research-ideation` skill + `literature-reviewer` agent + Zotero MCP **Process**: - **5W1H Brainstorming**: What, Why, Who, When, Where, How → structured thinking framework - **Literature Search & Import**: WebSearch finds papers → extract DOIs → auto-import to Zotero via `add_items_by_doi` → classify into themed sub-collections (Core Papers, Methods, Applications, Baselines, To-Read) - **PDF & Full-Text**: `find_and_attach_pdfs` batch-attaches open-access PDFs → `get_item_fulltext` reads full paper content for deep analysis (fallback: abstract + domain knowledge) - **Gap Analysis**: 5 types (Literature, Methodological, Application, Interdisciplinary, Temporal) → identify 2-3 concrete research opportunities - **Research Question**: SMART principles → formulate specific, measurable questions - **Method Selection & Planning**: Evaluate method applicability → timeline, milestones, risk assessment **Zotero Collection Structure**: ``` 📁 Research-{Topic}-{YYYY-MM} ├── 📁 Core Papers ├── 📁 Methods ├── 📁 Applications ├── 📁 Baselines └── 📁 To-Read ``` **Output**: `literature-review.md` + `research-proposal.md` + `references.bib` (exported from Zotero) + organized Zotero collection with PDFs **Commands**: - `/research-init "topic"` → full workflow: create Zotero collection → search & import papers → full-text analysis → gap analysis → generate review & proposal - `/zotero-review "collection"` → analyze an existing Zotero collection → generate literature review with comparison matrix - `/zotero-notes "collection"` → batch read papers → generate structured reading notes (summary/detailed/comparison formats) #### 2. ML Project Development Maintainable ML project structure for experiment code: **Tools**: `architecture-design` skill + `code-reviewer` agent + `git-workflow` skill **Process**: - **Structure**: Factory & Registry patterns → config-driven models (only `cfg` parameter) → enforced by `rules/coding-style.md` - **Code Style**: 200-400 line files → type hints required → `@dataclass(frozen=True)` for configs → max 3-level nesting - **Debug** (`bug-detective`): Error pattern matching for Python/Bash/JS → stack trace analysis → anti-pattern identification - **Git**: Conventional Commits (`feat/scope: message`) → branch strategy (master/develop/feature) → merge with `--no-ff` **Commands**: `/plan`, `/commit`, `/code-review`, `/tdd` #### 3. Experiment Analysis Statistical analysis and visualization of experimental results: **Tools**: `results-analysis` skill + `data-analyst` agent **Process**: - **Data Processing**: Automated cleaning and preprocessing of experiment logs - **Statistical Testing**: t-test, ANOVA, Wilcoxon signed-rank → validate significance - **Visualization**: matplotlib/seaborn integration → publication-ready figures (line plots, bar charts, heatmaps) - **Ablation Studies**: Systematic component analysis → understand contribution of each part **Command**: `/analyze-results <experiment_dir>` → generates analysis report with figures and statistics #### 4. Paper Writing Systematic paper writing from template to final draft: **Tools**: `ml-paper-writing` skill + `paper-miner` agent + `latex-conference-template-organizer` skill **Process**: - **Template Preparation**: Download conference .zip → extract main files → remove sample content → clean Overleaf-ready structure - **Citation Verification** (`citation-verification`): Multi-layer validation (Format → API → Information → Content) → prevents hallucinations - **Systematic Writing**: Narrative framing → 5-sentence abstract formula → section-by-section drafting with feedback cycles - **Anti-AI Processing** (`writing-anti-ai`): Remove inflated symbolism, promotional language, vague attributions → add human voice and rhythm → bilingual support (EN/CN) **Venues**: NeurIPS, ICML, ICLR, ACL, AAAI, COLM, Nature, Science, Cell, PNAS #### 5. Paper Self-Review Quality assurance before submission: **Tools**: `paper-self-review` skill **Process**: - **Structure Check**: Logical flow, section balance, narrative coherence - **Logic Validation**: Argument soundness, claim-evidence alignment, assumption clarity - **Citation Audit**: Reference accuracy, proper attribution, citation completeness - **Figure Quality**: Visual clarity, caption completeness, color accessibility - **Writing Polish**: Grammar, clarity, conciseness, academic tone - **Compliance**: Page limits, formatting requirements, ethical disclosures **6-item checklist** → systematic quality assessment #### 6. Submission & Rebuttal Paper submission and review response: **Tools**: `review-response` skill + `rebuttal-writer` agent **Submission Process**: - **Pre-submission**: Conference-specific checklists (NeurIPS 16-item, ICML Broader Impact, ICLR LLM disclosure) - **Format Check**: Page limits, anonymization, supplementary materials - **Final Review**: Proofread, check references, verify figures **Rebuttal Process**: - **Review Analysis**: Parse and classify comments (Major/Minor/Typo/Misunderstanding) - **Response Strategy**: Accept/Defend/Clarify/Experiment → tailored approach per comment type - **Rebuttal Writing**: Structured response with evidence and reasoning - **Tone Management**: Professional, respectful, evidence-based language **Command**: `/rebuttal <review_file>` → generates complete rebuttal document with experiment plan #### 7. Post-Acceptance Processing Conference preparation and research promotion: **Tools**: `post-acceptance` skill **Process**: - **Presentation**: Slide creation guidance (15/20/30 min formats) → visual design principles → storytelling structure - **Poster**: Academic poster templates (A0/A1 sizes) → layout optimization → visual hierarchy - **Promotion**: Social media content (Twitter/X, LinkedIn) → blog posts → press releases → research summaries **Commands**: `/presentation`, `/poster`, `/promote` → automated content generation **Coverage**: 90% of academic research lifecycle (from idea to publication) ### Supporting Workflows These workflows run in the background to enhance the primary workflows. #### Automated Enforcement Workflow Cross-platform hooks (Node.js) automate workflow enforcement: ``` Session Start → Skill Evaluation → Session End → Session Stop ``` - **skill-forced-eval** (`skill-forced-eval.js`): Before EVERY user prompt → groups all available skills (local + plugins) into 6 categories → silent scan mode, only outputs matched skills → requires activation before implementation → ensures no relevant skill is missed - **session-start** (`session-start.js`): Session begins → displays Git status, pending todos, available commands (top 5 with fold count), package manager → shows project context at a glance - **session-summary** (`session-summary.js`): Session ends → generates comprehensive work log → summarizes all changes made → provides smart recommendations for next steps → auto-cleans logs older than 30 days - **stop-summary** (`stop-summary.js`): Session stops → quick status check with separate added/modified/deleted counts → groups temp files by folder (top 3 per folder) → shows actionable cleanup suggestions - **security-guard** (`security-guard.js`): Two-tier security system — **Block tier**: immediately rejects catastrophic commands (rm -rf /, dd, mkfs, system dirs); **Confirm tier**: injects systemMessage forcing model to ask user before executing dangerous-but-legitimate operations (git push --force, git reset --hard, chmod 777, SQL DROP/DELETE/TRUNCATE, sensitive file writes) **Cross-platform**: All hooks use Node.js (not shell scripts) ensuring Windows/macOS/Linux compatibility. #### Knowledge Extraction Workflow Two specialized mining agents continuously extract knowledge to improve skills: - **paper-miner** (agent): Analyze research papers (PDF/DOCX/arXiv links) → extracts writing patterns, structure insights, venue requirements, rebuttal strategies → updates `ml-paper-writing/references/knowledge/` with categorized entries (structure.md, writing-techniques.md, submission-guides.md, review-response.md) - **kaggle-miner** (agent): Study winning Kaggle competition solutions → extract competition briefs, front-runner detailed technical analysis, code templates, best practices → update the `kaggle-learner` skill's knowledge base (`references/knowledge/[domain]/` directories, categorized by NLP/CV/Time Series/Tabular/Multimodal) **Knowledge feedback loop**: Each paper or solution analyzed enriches the knowledge base, creating a self-improving system that evolves with your research. #### Skill Evolution System 3-step continuous improvement cycle for maintaining and improving skills: ``` skill-development → skill-quality-reviewer → skill-improver ``` 1. **Develop** (`skill-development`): Create skills with proper YAML frontmatter → clear descriptions with trigger phrases → progressive disclosure (lean SKILL.md, details in `references/`) 2. **Review** (`skill-quality-reviewer`): 4-dimension quality assessment → Description Quality (25%), Content Organization (30%), Writing Style (20%), Structural Integrity (25%) → generates improvement plan with prioritized fixes 3. **Improve** (`skill-improver`): Merges suggested changes → updates documentation → iterates on feedback → reads improvement plans and applies changes automatically ## File Structure <details> <summary>View file structure</summary> ``` claude-scholar/ ├── hooks/ # Cross-platform JavaScript hooks (automated enforcement) │ ├── hook-common.js # Shared utilities (git diff, change analysis) │ ├── session-start.js # Session begin - Git status, todos, top 5 commands │ ├── skill-forced-eval.js # Silent scan, 6-category skill grouping │ ├── session-summary.js # Session end - work log, 30-day log auto-cleanup │ ├── stop-summary.js # Session stop - added/modified/deleted counts, grouped temp files │ └── security-guard.js # Two-tier security: Block (catastrophic) + Confirm (dangerous) │ ├── skills/ # 32 specialized skills (domain knowledge + workflows) │ ├── ml-paper-writing/ # Full paper writing: NeurIPS, ICML, ICLR, ACL, AAAI, COLM │ │ └── references/ │ │ └── knowledge/ # Extracted patterns from successful papers │ │ ├── structure.md # Paper organization patterns │ │ ├── writing-techniques.md # Sentence templates, transitions │ │ ├── submission-guides.md # Venue requirements (page limits, etc.) │ │ └── review-response.md # Rebuttal strategies │ │ │ ├── research-ideation/ # Research startup: 5W1H, literature review, gap analysis │ │ └── references/ │ │ ├── 5w1h-framework.md # Systematic thinking tool │ │ ├── gap-analysis-guide.md # 5 types of research gaps │ │ ├── literature-search-strategies.md │ │ ├── research-question-formulation.md │ │ ├── method-selection-guide.md │ │ └── research-planning.md │ │ │ ├── results-analysis/ # Experiment analysis: statistics, visualization, ablation │ │ └── references/ │ │ ├── statistical-methods.md # t-test, ANOVA, Wilcoxon │ │ ├── visualization-best-practices.md # matplotlib/seaborn │ │ ├── results-writing-guide.md # Writing results sections │ │ └── common-pitfalls.md # Common analysis mistakes │ │ │ ├── review-response/ # Systematic rebuttal writing │ │ └── references/ │ │ ├── review-classification.md # Major/Minor/Typo/Misunderstanding │ │ ├── response-strategies.md # Accept/Defend/Clarify/Experiment │ │ ├── rebuttal-templates.md # Structured response templates │ │ └── tone-guidelines.md # Professional language │ │ │ ├── paper-self-review/ # 6-item quality checklist │ ├── post-acceptance/ # Conference preparation │ │ └── references/ │ │ ├── presentation-templates/ # Slide creation (15/20/30 min) │ │ ├── poster-templates/ # Academic poster design │ │ ├── promotion-examples/ # Social media content │ │ └── design-guidelines.md # Visual design principles │ │ │ ├── citation-verification/ # Multi-layer citation validation │ ├── writing-anti-ai/ # Remove AI patterns: symbolism, promotional language │ │ └── references/ │ │ ├── patterns-english.md # English AI patterns to remove │ │ └── patterns-chinese.md # Chinese AI patterns to remove │ │ │ ├── architecture-design/ # ML project patterns: Factory, Registry, Config-driven │ ├── git-workflow/ # Git discipline: Conventional Commits, branching │ ├── bug-detective/ # Debugging: Python, Bash, JS/TS error patterns │ ├── code-review-excellence/ # Code review: security, performance, maintainability │ ├── skill-development/ # Skill creation: YAML, progressive disclosure │ ├── skill-quality-reviewer/ # Skill assessment: 4-dimension scoring │ ├── skill-improver/ # Skill evolution: merge improvements │ ├── kaggle-learner/ # Learn from Kaggle winning solutions │ ├── doc-coauthoring/ # Document collaboration workflow │ ├── latex-conference-template-organizer # Template cleanup for Overleaf │ └── ... (10+ more skills) │ ├── commands/ # 50+ slash commands (quick workflow execution) │ ├── research-init.md # Launch research startup workflow │ ├── zotero-review.md # Read Zotero papers, generate literature review │ ├── zotero-notes.md # Batch read Zotero papers, generate reading notes │ ├── analyze-results.md # Analyze experiment results │ ├── rebuttal.md # Generate systematic rebuttal document │ ├── presentation.md # Create conference presentation outline │ ├── poster.md # Generate academic poster design plan │ ├── promote.md # Generate promotion content │ ├── plan.md # Implementation planning with agent delegation │ ├── commit.md # Conventional Commits: feat/fix/docs/refactor │ ├── code-review.md # Quality and security review workflow │ ├── tdd.md # Test-driven development: Red-Green-Refactor │ ├── build-fix.md # Fix build errors automatically │ ├── verify.md # Run verification loops │ ├── checkpoint.md # Save verification state │ ├── refactor-clean.md # Remove dead code │ ├── learn.md # Extract patterns from code │ ├── update-github.md # Commit and push to GitHub │ ├── update-readme.md # Update README documentation │ ├── update-memory.md # Check and update CLAUDE.md memory │ ├── create_project.md # Create new project from template │ ├── setup-pm.md # Configure package manager (uv/pnpm) │ └── sc/ # SuperClaude command suite (30 commands) │ ├── sc-agent.md # Agent management │ ├── sc-estimate.md # Development time estimation │ ├── sc-improve.md # Code improvement │ └── ... │ ├── agents/ # 14 specialized agents (focused task delegation) │ ├── literature-reviewer.md # Literature search and trend analysis │ ├── data-analyst.md # Automated data analysis and visualization │ ├── rebuttal-writer.md # Systematic rebuttal writing │ ├── paper-miner.md # Extract paper knowledge: structure, techniques │ ├── architect.md # System design: architecture decisions │ ├── code-reviewer.md # Review code: quality, security, best practices │ ├── tdd-guide.md # Guide TDD: test-first development │ ├── kaggle-miner.md # Extract engineering practices from Kaggle │ ├── build-error-resolver.md # Fix build errors: analyze and resolve │ ├── refactor-cleaner.md # Remove dead code: detect and cleanup │ ├── bug-analyzer.md # Deep code execution flow analysis and root cause investigation │ ├── dev-planner.md # Implementation planning and task breakdown │ ├── ui-sketcher.md # UI blueprint design and interaction specs │ └── story-generator.md # User story and requirement generation │ ├── rules/ # Global guidelines (always-follow constraints) │ ├── coding-style.md # ML project standards: file size, immutability, types │ ├── agents.md # Agent orchestration: when to delegate, parallel execution │ ├── security.md # Secrets management, sensitive file protection │ └── experiment-reproducibility.md # Random seeds, config recording, checkpoints │ ├── CLAUDE.md # Global configuration: project overview, preferences, rules │ └── README.md # This file - overview, installation, features ``` </details> ## Feature Highlights ### Skills (32 total) **Web Design:** - `frontend-design` - Create distinctive, production-grade frontend interfaces - `ui-ux-pro-max` - UI/UX design intelligence (50+ styles, 97 palettes, 9 stacks) - `web-design-reviewer` - Visual inspection and design issue fixing **Writing & Academic:** - `ml-paper-writing` - Full paper writing guidance for top conferences/journals - `writing-anti-ai` - Remove AI writing patterns (bilingual support) - `doc-coauthoring` - Structured document collaboration workflow - `latex-conference-template-organizer` - LaTeX template management - `daily-paper-generator` - Automated daily paper generation for research tracking **Research Workflow:** - `research-ideation` - Research startup: 5W1H brainstorming, literature review, gap analysis - `results-analysis` - Experiment analysis: statistical testing, visualization, ablation studies - `review-response` - Systematic rebuttal writing with tone management - `paper-self-review` - 6-item quality checklist for paper self-assessment - `post-acceptance` - Conference preparation: presentations, posters, promotion - `citation-verification` - Multi-layer citation validation to prevent hallucinations **Development:** - `daily-coding` - Daily coding checklist (minimal, auto-triggered) - `git-workflow` - Git best practices (Conventional Commits, branching) - `code-review-excellence` - Code review guidelines - `bug-detective` - Debugging for Python, Bash, JS/TS - `architecture-design` - ML project design patterns - `verification-loop` - Testing and validation **Plugin Development:** - `skill-development` - Skill creation guide - `skill-improver` - Skill improvement tools - `skill-quality-reviewer` - Quality assessment - `command-development` - Slash command creation - `agent-identifier` - Agent configuration - `hook-development` - Hook development guide - `mcp-integration` - MCP server integration **Utilities:** - `uv-package-manager` - Modern Python package management - `planning-with-files` - Markdown-based planning - `webapp-testing` - Local web application testing - `kaggle-learner` - Learn from Kaggle solutions ### Commands (50+) **Research Commands:** | Command | Purpose | |---------|---------| | `/research-init` | Launch research startup workflow (5W1H, literature review, gap analysis) | | `/zotero-review` | Read papers from Zotero collection, generate structured literature review | | `/zotero-notes` | Batch read Zotero papers, generate structured reading notes | | `/analyze-results` | Analyze experiment results (statistics, visualization, ablation) | | `/rebuttal` | Generate systematic rebuttal document from review comments | | `/presentation` | Create conference presentation outline | | `/poster` | Generate academic poster design plan | | `/promote` | Generate promotion content (Twitter, LinkedIn, blog) | **Development Commands:** | Command | Purpose | |---------|---------| | `/plan` | Create implementation plans | | `/commit` | Commit with Conventional Commits | | `/update-github` | Commit and push to GitHub | | `/update-readme` | Update README documentation | | `/update-memory` | Check and update CLAUDE.md memory | | `/code-review` | Perform code review | | `/tdd` | Test-driven development workflow | | `/build-fix` | Fix build errors | | `/verify` | Verify changes | | `/checkpoint` | Create checkpoints | | `/refactor-clean` | Refactor and cleanup | | `/learn` | Extract reusable patterns | | `/create_project` | Create new project from template | | `/setup-pm` | Configure package manager (uv/pnpm) | | `/sc` | SuperClaude command suite (30 commands) | ### Agents (14 specialized) **Research Agents:** - **literature-reviewer** - Literature search, classification, and trend analysis - **data-analyst** - Automated data analysis and visualization - **rebuttal-writer** - Systematic rebuttal writing with tone optimization - **paper-miner** - Extract paper writing knowledge from successful publications **Development Agents:** - **architect** - System architecture design - **build-error-resolver** - Fix build errors - **code-reviewer** - Review code quality - **refactor-cleaner** - Remove dead code - **tdd-guide** - Guide TDD workflow - **kaggle-miner** - Extract Kaggle engineering practices - **bug-analyzer** - Deep code execution flow analysis and root cause investigation - **dev-planner** - Implementation planning and task breakdown **Design & Content Agents:** - **ui-sketcher** - UI blueprint design and interaction specs - **story-generator** - User story and requirement generation ## Quick Start ### Installation Options Choose the installation method that fits your needs: #### Option 1: Full Installation (Recommended) ```bash git clone https://github.com/Galaxy-Dawn/claude-scholar.git /tmp/claude-scholar bash /tmp/claude-scholar/scripts/setup.sh ``` The script merges skills/commands/agents/rules/hooks into your existing `~/.claude`, and adds hooks/mcpServers/enabledPlugins to your `settings.json` (auto-backup to `settings.json.bak`). Your env and permissions are untouched. **Includes**: All 32 skills, 50+ commands, 14 agents, 5 hooks, and project rules. #### Option 2: Minimal Installation Core hooks and essential skills only (faster load, less complexity): ```bash # Clone repository git clone https://github.com/Galaxy-Dawn/claude-scholar.git /tmp/claude-scholar # Copy only hooks and core skills mkdir -p ~/.claude/hooks ~/.claude/skills cp /tmp/claude-scholar/hooks/*.js ~/.claude/hooks/ cp -r /tmp/claude-scholar/skills/ml-paper-writing ~/.claude/skills/ cp -r /tmp/claude-scholar/skills/research-ideation ~/.claude/skills/ cp -r /tmp/claude-scholar/skills/results-analysis ~/.claude/skills/ cp -r /tmp/claude-scholar/skills/review-response ~/.claude/skills/ cp -r /tmp/claude-scholar/skills/writing-anti-ai ~/.claude/skills/ cp -r /tmp/claude-scholar/skills/git-workflow ~/.claude/skills/ cp -r /tmp/claude-scholar/skills/bug-detective ~/.claude/skills/ # Cleanup rm -rf /tmp/claude-scholar ``` **Post-install**: Merge hooks config into your `settings.json` — see `settings.json.template` for the required hooks entries. **Includes**: 5 hooks, 7 core skills (complete research workflow + essential development). #### Option 3: Selective Installation Pick and choose specific components: ```bash # Clone repository git clone https://github.com/Galaxy-Dawn/claude-scholar.git /tmp/claude-scholar cd /tmp/claude-scholar # Copy what you need, for example: # - Hooks only cp hooks/*.js ~/.claude/hooks/ # - Specific skills cp -r skills/latex-conference-template-organizer ~/.claude/skills/ cp -r skills/architecture-design ~/.claude/skills/ # - Specific agents cp agents/paper-miner.md ~/.claude/agents/ # - Project rules cp rules/coding-style.md ~/.claude/rules/ cp rules/agents.md ~/.claude/rules/ ``` **Post-install**: Merge hooks config into your `settings.json` — see `settings.json.template`. **Recommended for**: Advanced users who want custom configurations. ### Requirements - Claude Code CLI - Git - Node.js (required for hooks) - (Optional) uv, Python (for Python development) - **Zotero desktop app** (for Zotero MCP features) ### MCP Setup (Optional) For Zotero-integrated research workflows, install the MCP server: ```bash # Install Zotero MCP server uv tool install zotero-mcp-server # Enable Local API in Zotero desktop app: # Edit → Settings → Advanced → Check "Allow other applications on this computer to communicate with Zotero" ``` Then add to your `~/.claude/settings.json`: ```json { "mcpServers": { "zotero": { "command": "zotero-mcp", "args": ["serve"], "env": { "ZOTERO_LOCAL": "true", "NO_PROXY": "localhost,127.0.0.1" } } } } ``` See [MCP_SETUP.md](./MCP_SETUP.md) for detailed setup guide and troubleshooting. ### First Run After installation, the hooks provide automated workflow assistance: 1. **Every prompt** triggers `skill-forced-eval` → ensures applicable skills are considered 2. **Session starts** with `session-start` → displays project context 3. **Sessions end** with `session-summary` → generates work log with recommendations 4. **Session stops** with `stop-summary` → provides status check ## Project Rules ### Coding Style Enforced by `rules/coding-style.md`: - **File Size**: 200-400 lines maximum - **Immutability**: Use `@dataclass(frozen=True)` for configs - **Type Hints**: Required for all functions - **Patterns**: Factory & Registry for all modules - **Config-Driven**: Models accept only `cfg` parameter ### Agent Orchestration Defined in `rules/agents.md`: - Available agent types and purposes - Parallel task execution - Multi-perspective analysis ### Security Defined in `rules/security.md`: - Secrets management (environment variables, `.env` files) - Sensitive file protection (never commit tokens, keys, credentials) - Pre-commit security checks via hooks ### Experiment Reproducibility Defined in `rules/experiment-reproducibility.md`: - Random seed management for reproducibility - Configuration recording (Hydra auto-save) - Environment recording and checkpoint management ## Contributing This is a personal configuration, but you're welcome to: - Fork and adapt for your own research - Submit issues for bugs - Suggest improvements via issues ## License MIT License ## Acknowledgments Built with Claude Code CLI and enhanced by the open-source community. ### References This project is inspired by and builds upon excellent work from the community: - **[everything-claude-code](https://github.com/anthropics/everything-claude-code)** - Comprehensive resource for Claude Code CLI - **[AI-research-SKILLs](https://github.com/zechenzhangAGI/AI-research-SKILLs)** - Research-focused skills and configurations These projects provided valuable insights and foundations for the research-oriented features in Claude Scholar. --- **For data science, AI research, and academic writing.** Repository: [https://github.com/Galaxy-Dawn/claude-scholar](https://github.com/Galaxy-Dawn/claude-scholar)