[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-84006":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":12,"contributorsCount":12,"subscribersCount":12,"size":12,"stars1d":14,"stars7d":15,"stars30d":15,"stars90d":12,"forks30d":12,"starsTrendScore":16,"compositeScore":12,"rankGlobal":9,"rankLanguage":9,"license":17,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":20,"hasPages":18,"topics":21,"createdAt":9,"pushedAt":9,"updatedAt":22,"readmeContent":23,"aiSummary":9,"trendingCount":12,"starSnapshotCount":12,"syncStatus":14,"lastSyncTime":24,"discoverSource":25},84006,"Literature-Mind","meishiwhy\u002FLiterature-Mind","meishiwhy","LitMind 将你的 Zotero 文献库转化为一个能够检索、推理和辅助写作的 AI 科研知识库。",null,"Python",69,0,51,2,16,20,"MIT License",false,"main",true,[],"2026-06-12 02:04:37","# LitMind\n\n学术文献智能处理工具集。覆盖文献从元数据到结构化知识的全流程，帮助科研人员自动完成文献检索、PDF 解析、知识提取、证据检索、Discussion 生成和综述撰写。\n\n## 模块总览\n\n```\nZotero → Metadata → PDF → Full Text → Analysis → KB → Evidence → Discussion → Review\n (m1)      (m2)      (m3)         (m4\u002F5)     (m6)      (m7)        (m8)\n```\n\n| 模块 | 名称 | 功能 |\n|---|---|---|\n| m1 | **litmind-zotero** | Zotero 连接器 — 导出文献元数据 |\n| m2 | **litmind-parser** | PDF 解析 — 清洗噪声、识别章节结构 |\n| m3 | **litmind-analyzer** | LLM 知识提取 — 研究问题\u002F方法\u002F变量\u002F发现\u002F声明 |\n| m4 | **litmind-knowledge** | 知识库 — SQLite + ChromaDB 存储与检索 |\n| m5 | **litmind-chat** | 科研问答 — 基于知识库的自然语言问答 |\n| m6 | **litmind-evidence** | 证据检索 — 支持\u002F反对\u002F中性证据分类 |\n| m7 | **litmind-discussion** | Discussion 生成 — 7 节结构化草稿 |\n| m8 | **litmind-review** | 综述生成 — 主题聚类\u002F共识\u002F争议\u002F空白+全文 |\n\n---\n\n## 快速开始\n\n### 安装\n\n```bash\npip install litmind\n\n# 按需安装依赖\npip install litmind[pdf]     # PDF 解析 (pymupdf)\npip install litmind[llm]     # LLM 分析 (anthropic + openai)\npip install litmind[kb]      # 知识库 (sqlalchemy + chromadb)\npip install litmind[all]     # 全部依赖\n```\n\n### 环境变量\n\n```bash\nexport ANTHROPIC_API_KEY=sk-...   # Claude provider (m3, m5, m6, m7, m8)\nexport OPENAI_API_KEY=sk-...      # OpenAI provider (可选)\n```\n\n---\n\n## 模块详解\n\n### m1: Zotero Connector (`\u002Flitmind-zotero`)\n\n从 Zotero 本地 SQLite 数据库读取期刊论文元数据。\n\n```bash\npython scripts\u002Fcli.py export -o papers.json\npython scripts\u002Fcli.py stats\n```\n\n```python\nfrom litmind_zotero import discover_database, export_all\npapers = export_all(discover_database())\n```\n\n**输出：** `PaperMetadata` — key, title, authors, year, doi, journal, pdfPath, tags, collections\n\n---\n\n### m2: Paper Parser (`\u002Flitmind-parser`)\n\n从 PDF 提取全文 → 自动清洗（页眉\u002F页脚\u002F页码\u002F重复）→ 识别标准章节。\n\n```bash\npython scripts\u002Fparse.py single paper.pdf -o parsed.json\npython scripts\u002Fparse.py batch --from-zotero papers.json -o parsed\u002F\n```\n\n```python\nfrom litmind_parser import parse_pdf\nresult = parse_pdf(\"paper.pdf\")\nprint(result.sections.abstract[:200])\n```\n\n**输出：** `PaperContent` — fullText + sections (abstract\u002Fintro\u002Fmethods\u002Fresults\u002Fdiscussion\u002Fconclusion)\n\n---\n\n### m3: Paper Analyzer (`\u002Flitmind-analyzer`)\n\n将论文全文通过 LLM 提取为结构化科研知识。\n\n```bash\nlitmind-analyze parsed.json -o analysis.json\nlitmind-analyze paper.json --provider openai --model gpt-4o\n```\n\n```python\nfrom litmind_analyzer import analyze_paper\nfrom litmind_analyzer.providers import AnthropicProvider\n\nprovider = AnthropicProvider()\nresult = analyze_paper(paper_content, provider)\nprint(f\"研究问题: {result.researchQuestion}\")\nprint(f\"方法: {result.methods}\")\n```\n\n**输出：** `PaperAnalysis` — researchQuestion, studyDesign, participants, methods, statistics, variables, claims, limitations, keywords\n\n---\n\n### m4: Knowledge Base (`\u002Flitmind-knowledge`)\n\n基于 SQLite + ChromaDB 的本地科研知识库，存储论文分析结果，支持语义检索和结构化查询。\n\n```bash\nlitmind-knowledge add analysis.json\nlitmind-knowledge search \"flatfoot kinematics\"\nlitmind-knowledge stats\n```\n\n```python\nfrom litmind_knowledge.service import KnowledgeBase\n\nkb = KnowledgeBase()\nkb.add_paper(analysis_dict)\nresults = kb.semantic_search(\"MTP ROM flatfoot\", top_k=10)\n```\n\n---\n\n### m5: Research Chat (`\u002Flitmind-chat`)\n\n面向知识库的自然语言问答。自动分类问题类型、检索相关证据、生成带引用的回答。\n\n```bash\nlitmind-chat \"What studies support the link between flatfoot and MTP ROM?\"\nlitmind-chat \"Show me papers using SPM analysis\"\n```\n\n```python\nfrom litmind_chat.service import ResearchChatService\n\nchat = ResearchChatService(kb=kb, llm_provider=provider)\nresponse = chat.ask(\"Does flatfoot increase forefoot motion?\")\nprint(f\"Answer: {response.answer}\")\nprint(f\"Sources: {len(response.supportingPapers)} papers\")\n```\n\n---\n\n### m6: Evidence Finder (`\u002Flitmind-evidence`)\n\n输入一个科学观点，自动检索知识库中的支持证据、反对证据和中性证据，评估证据强度。\n\n```bash\npython scripts\u002Fevidence.py \"Flatfoot increases MTP ROM\"\npython scripts\u002Fevidence.py \"Carbon plate shoes improve jump performance\" --json\n```\n\n```python\nfrom litmind_evidence import EvidenceFinderService\n\nev_service = EvidenceFinderService(kb=kb, llm_provider=provider)\nresult = ev_service.find_evidence(\"SPM is more sensitive than peak-value analysis\")\nprint(f\"Strength: {result.evidenceStrength}\")  # Strongly \u002F Moderately \u002F Weakly Supported\nprint(f\"Support: {len(result.support)} papers\")\nprint(f\"Oppose: {len(result.oppose)} papers\")\n```\n\n**证据强度分级：** Strongly Supported | Moderately Supported | Weakly Supported | Mixed Evidence | Insufficient Evidence\n\n---\n\n### m7: Discussion Generator (`\u002Flitmind-discussion`)\n\n输入研究结果，自动检索相关文献，生成 7 节结构化 Discussion 草稿。\n\n```bash\npython scripts\u002Fdiscussion.py \\\n  --topic \"Footwear stiffness effects on biomechanics\" \\\n  --results \"High stiffness shoes increased MTP ROM\" \\\n  --results \"No significant difference in ankle sagittal ROM\"\n```\n\n```python\nfrom litmind_discussion import DiscussionGeneratorService, DiscussionInput\n\nservice = DiscussionGeneratorService(evidence_service=ev_service, llm_provider=provider)\ninp = DiscussionInput(studyTopic=\"Footwear stiffness\", results=[\"...\", \"...\"])\nresult = service.generate_discussion(inp)\nprint(result.discussionDraft)\n```\n\n**7 个 Section：**\n1. Main Finding Interpretation\n2. Supporting Evidence\n3. Contradictory Evidence\n4. Potential Mechanisms\n5. Practical Implications\n6. Study Limitations\n7. Future Directions\n\n---\n\n### m8: Review Generator (`\u002Flitmind-review`)\n\n输入一个研究主题，自动完成文献发现、主题聚类、趋势分析、共识与争议识别、研究空白发现，生成结构化综述全文。\n\n```bash\npython scripts\u002Freview.py \"Flatfoot Biomechanics\"\npython scripts\u002Freview.py \"SPM in Biomechanics\" --json\n```\n\n```python\nfrom litmind_review import ReviewGeneratorService, ReviewInput\n\nservice = ReviewGeneratorService(kb=kb, evidence_service=ev_service, llm_provider=provider)\ninp = ReviewInput(topic=\"Flatfoot Biomechanics\", max_papers=50)\nresult = service.generate_review(inp)\n\nprint(f\"Themes: {len(result.researchThemes)}\")\nprint(f\"Consensus: {len(result.researchConsensus)}\")\nprint(f\"Controversies: {len(result.researchControversies)}\")\nprint(f\"Gaps: {len(result.researchGaps)}\")\nprint(result.reviewDraft)\n```\n\n**核心功能：**\n- **ThemeDiscoveryEngine** — LLM 自动聚类，提炼 3-7 个研究主题\n- **TrendAnalyzer** — 统计高频变量、统计方法、研究设计、年份分布\n- **ConsensusAnalyzer** — 识别多篇文献一致支持的结论\n- **ControversyAnalyzer** — 发现支持和反对证据并存的争议点\n- **GapAnalyzer** — 找出低频研究方向和空白领域\n- **ReviewComposer** — 逐节生成 8 个 Section 的综述全文\n\n**8 个 Section：**\n1. Introduction\n2. Current Research Landscape\n3. Major Research Themes\n4. Evidence Consensus\n5. Research Controversies\n6. Research Gaps\n7. Future Directions\n8. Conclusion\n\n---\n\n## 引用安全\n\n所有模块的 LLM 生成内容均遵循严格的白名单引用机制：\n\n- LLM 只能引用 Knowledge Base 中的真实 paperId\n- 所有引用在后处理阶段校验，未通过的自动丢弃\n- 禁止虚构作者、年份、DOI、期刊\n\n---\n\n## Claude Code Skill\n\n在 Claude Code 中可直接调用以下命令：\n\n```\n\u002Flitmind-zotero      Zotero 连接器\n\u002Flitmind-parser      PDF 解析\n\u002Flitmind-analyzer    论文知识提取\n\u002Flitmind-knowledge   知识库\n\u002Flitmind-chat        科研问答\n\u002Flitmind-evidence    证据检索\n\u002Flitmind-discussion  Discussion 生成\n\u002Flitmind-review      综述生成\n```\n\n---\n\n## 项目结构\n\n```\nlitmind\u002F\n├── src\u002F\n│   ├── litmind_zotero\u002F         # m1\n│   ├── litmind_parser\u002F         # m2\n│   ├── litmind_analyzer\u002F       # m3\n│   ├── litmind_knowledge\u002F      # m4\n│   ├── litmind_chat\u002F           # m5\n│   ├── litmind_evidence\u002F       # m6\n│   ├── litmind_discussion\u002F     # m7\n│   └── litmind_review\u002F         # m8\n├── scripts\u002F\n├── tests\u002F\n├── .claude\u002Fskills\u002F\n├── pyproject.toml\n└── README.md\n```\n\n---\n\n## License\n\nMIT\n","2026-06-11 04:12:04","CREATED_QUERY"]