citationclaw 1.0.6

论文被引画像分析工具 — 自动爬取施引文献、识别著名学者、生成可视化 HTML 报告

English | 中文

CitationClaw: A Lightweight Engine for Discovering Scientific Impact through Citations

让每一次引用都成为可解释的影响力
Turning Every Citation into Explainable Impact

Turn Every Citation into Explainable Impact.
Input paper titles (or import from Google Scholar profiles), and generate a full citation portrait report in minutes.

🚀 Contribute with PRs

CitationClaw is community-driven and PR-friendly.

• Open an issue: https://github.com/VisionXLab/CitationClaw/issues
• Submit a PR: https://github.com/VisionXLab/CitationClaw/pulls
• Good first tasks: docs, UI polish, skill metadata, retry robustness

📢 News

• 2026-03-15: Released beta v1.0.6 — English README as default, Chinese switch at top, and usage flow linked to Guidelines Quick Start.
• 2026-03-14: Released v1.0.5 — AI assistant widgets for UI/report pages and reliability fixes.
• 2026-03-14: Released v1.0.4 — improved UI and introduced Basic/Advanced/Full service tiers.
• 2026-03-12: Released v1.0 — first public release.

Key Features

• 🧠 Five-Phase Citation Pipeline: crawl -> author intelligence -> export -> citing description -> dashboard.
• 🎯 Renowned Scholar Focus: auto-identifies high-impact scholars and generates dedicated outputs.
• ⚡ Tiered Analysis Modes: Basic / Advanced / Full for speed-cost-depth tradeoff.
• 🔁 Resumable + Cache-Aware: supports resume-by-page, author cache, and citing-description cache.
• 📊 Shareable HTML Report: standalone dashboard file, no extra server needed for viewing.
• 🧩 Skills Runtime Inside: keeps five-phase logic while moving execution to modular skills.

🏗️ Architecture

CitationClaw keeps deterministic business phases while using a skills-style runtime for orchestration.

UI/REST/WebSocket
      │
      ▼
TaskExecutor (Orchestrator)
      │
      ▼
Skills Runtime
  ├─ phase1_citation_fetch
  ├─ phase2_author_intel
  ├─ phase3_export
  ├─ phase4_citation_desc
  └─ phase5_report_generate

More details: Technical Report

Table of Contents

• News
• Key Features
• Architecture
• Install
• Quick Start
• Configuration Highlights
• Project Structure
• Outputs
• Contribute & Roadmap
• Community
• Star History
• Disclaimer

📦 Install

Requires Python 3.10+ (Python 3.12 recommended).

Install from PyPI (recommended)

pip install citationclaw
citationclaw                  # default: 127.0.0.1:8000
citationclaw --port 8080      # custom port

Install from source

git clone https://github.com/VisionXLab/CitationClaw.git
cd CitationClaw
pip install -r requirements.txt
python start.py               # default: 127.0.0.1:8000
python start.py --port 8080

🚀 Quick Start

For first-time users, follow the complete guide with screenshots:

• Quick Start (online)
• Quick Start (local file)

⚙️ Configuration Highlights

• Required keys:
  • ScraperAPI Key(s) for Google Scholar crawling
  • OpenAI-compatible API Key for LLM-based analysis
• Recommended search model:
  • Keep gemini-3-flash-preview-search for search-capable stages
• Service tiers:
  • Basic: lower cost and faster for first runs
  • Advanced: citing descriptions for renowned-scholar papers only
  • Full: citing descriptions for all citing papers
• For papers with >1000 citations:
  • Enable year traverse mode

📁 Project Structure

citationclaw/
├── app/                 # FastAPI app, task orchestration, config, logs
├── core/                # scraping / search / export / dashboard engines
├── skills/              # skills runtime and five phase skills
├── static/              # frontend assets
├── templates/           # Jinja2 pages
docs/                    # docs and demos
test/                    # tests

📤 Outputs

Each run creates a timestamped folder under data/result-{timestamp}/, usually including:

• paper_results.xlsx
• paper_results_all_renowned_scholar.xlsx
• paper_results_top-tier_scholar.xlsx
• paper_results_with_citing_desc.xlsx
• paper_results.json
• paper_dashboard.html

🤝 Contribute & Roadmap

PRs are welcome and appreciated.

Suggested directions:

• richer skill metadata and registry conventions
• stronger retry and network-failure resilience
• dashboard readability and UX improvement
• tests for pipeline contracts and compatibility
• provider/model compatibility presets

Useful links:

• Issues: https://github.com/VisionXLab/CitationClaw/issues
• Pull Requests: https://github.com/VisionXLab/CitationClaw/pulls
• Guidelines: https://visionxlab.github.io/CitationClaw/guidelines.html

🌍 Community

• Product update: 减论 reduct.cn
• User group (CN):

⭐ Star History