MathModel Figure Templates
jihe520/MathModelAgent
Renders ready-made scientific figure templates, such as raincloud plots, Taylor diagrams and chord diagrams, from bundled Python scripts in the MathModel sandbox.
Academic integrity investigation and scholar profile analysis system.
Install Academic Investigation
$ npx skills add yadnuses/Academic-Detective --skill academic-investigation -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install yadnuses/Academic-Detective academic-investigation --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
Claude Code skills documentation · loads skills from .claude/skills/
Install the "academic-investigation" agent skill from https://github.com/yadnuses/Academic-Detective/tree/main into .claude/skills/academic-investigation/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "academic-investigation", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add yadnuses/Academic-Detective --skill academic-investigation -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install yadnuses/Academic-Detective academic-investigation --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "academic-investigation" agent skill from https://github.com/yadnuses/Academic-Detective/tree/main into .agents/skills/academic-investigation/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "academic-investigation", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add yadnuses/Academic-Detective --skill academic-investigation -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install yadnuses/Academic-Detective academic-investigation --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "academic-investigation" agent skill from https://github.com/yadnuses/Academic-Detective/tree/main into .cursor/skills/academic-investigation/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "academic-investigation", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add yadnuses/Academic-Detective --skill academic-investigation -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install yadnuses/Academic-Detective academic-investigation --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "academic-investigation" agent skill from https://github.com/yadnuses/Academic-Detective/tree/main into .gemini/skills/academic-investigation/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "academic-investigation", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install yadnuses/Academic-Detective academic-investigationInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add yadnuses/Academic-Detective --skill academic-investigation -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "academic-investigation" agent skill from https://github.com/yadnuses/Academic-Detective/tree/main into .github/skills/academic-investigation/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "academic-investigation", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add yadnuses/Academic-Detective --skill academic-investigation -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install yadnuses/Academic-Detective academic-investigation --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "academic-investigation" agent skill from https://github.com/yadnuses/Academic-Detective/tree/main into .opencode/skills/academic-investigation/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "academic-investigation", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
academic-investigation- GitHub stars
- 313
- Token cost
- ~4.2k tokens
- SKILL.md length
- 1,223 words
- Files
- 345 (incl. scripts)
- Skills in repo
- 1
- Repo updated
- First seen
- Licence
- MIT
At a glance
Academic integrity investigation and scholar profile analysis system.
- Works in 3 steps: 提取引用 → LLM 逐条验证 → 生成报告
- Conducting comprehensive academic background checks
- SKILL.md covers Overview, When to Use, Semi-Automatic Workflow… and 精准工具索引表, plus 5 more sections
- Calls python
What it does
Academic Investigation is an agent skill from yadnuses/Academic-Detective. Academic integrity investigation and scholar profile analysis system. Use when conducting comprehensive academic background checks, credential verification, publication analysis, or institutional affiliation audits of scholars, researchers, or faculty members. Triggers on phrases like "调查学者", "学术调查", "查某人的学术背景", "核实论文", "学术档案调查", "学者背调", or when analyzing academic misconduct, credential inflation, publication quality, or research integrity issues.
Its SKILL.md is about 4.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 351 other files, including scripts (for example `.github/ISSUE_TEMPLATE/apply_for_investigation.md`, `.github/ISSUE_TEMPLATE/claim_investigation.md` and `.github/PULL_REQUEST_TEMPLATE/contribute_case.md`).
It sits in Data & Analytics, covering Data visualization. The repository describes itself as: +Academic Detective · 学术侦探 — 开源学术背景核查引擎。输入导师姓名和学校，基于公开数据自动生成调查报告：论文产出核实、六维质量评分、学科基准线偏差检测、与已知不端案例的模式比对、结构化学生评价匹配、关系网络可视化。支持 Markdown→PDF 报告导出与导师知识蒸馏。 The licence is MIT.
When your agent uses it
- Conducting comprehensive academic background checks
- Credential verification
- Publication analysis
- Institutional affiliation audits of scholars
Example prompts
- “查某人的学术背景”
- “学术档案调查”
- “/academic-investigation”
Requirements
- Python 3
Workflow steps
3 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit 8de7b96. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
python
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Academic Investigation loads about 4.2k tokens when it runs. Until then it costs about 119 tokens; SKILL.md has 1,223 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from yadnuses/Academic-Detective at commit 8de7b96, republished under its MIT licence (© yadnuses). 1,223 words, ~4,217 tokens.
Download SKILL.mdSave it as
.claude/skills/academic-investigation/SKILL.md (or your agent's skills folder). This skill also uses 344 other files; get the full folder from GitHub.- name
- academic-investigation
- description
- Academic integrity investigation and scholar profile analysis system. Use when conducting comprehensive academic background checks, credential verification, publication analysis, or institutional affiliation audits of scholars, researchers, or faculty members. Triggers on phrases like "调查学者", "学术调查", "查某人的学术背景", "核实论文", "学术档案调查", "学者背调", or when analyzing academic misconduct, credential inflation, publication quality, or research integrity issues.
⚠️ LLM 强制指令（请逐条执行）：
- 优先使用现有脚本。本系统的工具链位于
scripts/目录下，一般情况下不要自己写 Python 代码替代现有工具。- 所有脚本路径以下方工具索引表为准。
scripts/根目录下绝大多数 <500 字节的.py文件是兼容性 shim，真实实现位于子目录中。- 执行流程：先 ls 确认文件存在 → 再执行。如果不确定某个脚本的具体路径或参数，先 Read 该脚本头部的 docstring。
- 遵循七步顺序：Basic Profile → Output Quantity → Quality Assessment → Relationship Network → Anomaly Detection → Multi-Source Validation → Report Generation。不要跳步。
- 分层加载：执行具体步骤前，先 Read 对应的
docs/skill/文档获取详细流程。- 自由边界：对于客户要求但没有在目录写出的功能，可由LLM编写脚本或自主筛查以保证功能实现。
Academic Investigation Skill
Comprehensive academic background investigation system based on proven methodology from verified case studies.
Overview
A systematic 7-step framework for investigating academic profiles, verifying credentials, analyzing publication quality, and identifying potential academic misconduct or credential inflation.
When to Use
- Investigating a scholar's academic background or credentials
- Verifying publication claims or academic achievements
- Analyzing research quality and originality
- Checking for academic misconduct (plagiarism, duplicate publication, credential fraud)
- Auditing institutional affiliations and career progression
- Evaluating relationship networks and resource dependencies
Semi-Automatic Workflow Philosophy
Human-in-the-loop design. Scripts assist with computation; human judgment drives decisions.
| Actor | Responsibilities |
|---|---|
| Human | CNKI/Wanfang/WoS searches, institutional verification, monograph acquisition, final interpretation |
| LLM | Tool selection, signal interpretation, hypothesis generation, report drafting |
| Scripts | Data normalization, quantitative analysis, text profiling, report assembly |
scripts/
├── core/ # 共享基础设施
├── domestic/ # 国内学者调查适配器
├── international/ # 国际学者/导师调查适配器
├── cross_border/ # 海归学者调查（合并国内+国际）
├── analysis/ # 共享分析模块
├── network/ # 关系网络与腐败图谱
├── deep_evidence/ # 深度证据层（数据取证/发表链/伦理/同行评议/证据编译）
├── report/ # 报告生成
├── agents/ # 多智能体协作层
├── delivery/ # 交付层（自检+格式化）
├── backend/ # Web 后端 API
└── investigate.py # CLI 编排入口精准工具索引表
每个工具的真实路径已排除根目录兼容性 shim。执行前先
ls确认文件存在。
Core（基础设施）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track |
|---|---|---|---|---|---|---|
| case_manager | scripts/core/case_manager.py | 案件注册与 ID 生成 (AD-YYYY-MM-DD-NNN) | 甲方名称 | case_id + 目录结构 | Step 0 | all |
| db | scripts/core/db.py | SQLite 案件数据库（9表） | case_id | .db 文件 | Step 0 后自动 | all |
| config_loader | scripts/core/config_loader.py | 统一配置加载（v1→v2 迁移） | config.yaml | config dict | 全程 | all |
| router | scripts/core/router.py | 调查类型路由 (domestic/international/cross_border) | config | track 判定 | Step 0 | all |
| watermark | scripts/core/watermark.py | 零宽水印嵌入/提取 | Markdown 报告 | 水印报告 | Step 8 交付 | all |
| utils | scripts/core/utils.py | 日志、JSON 存储等公共工具 | — | — | 全程 | all |
Data Import（数据导入）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| data_importer | scripts/domestic/data_importer.py | CNKI/万方/WoS 导入 + 去重 | 导出的 txt/csv 文件 | 统一 JSON 论文列表 | Step 2 | domestic | init 完成 |
| data_fetcher | scripts/international/data_fetcher.py | OpenAlex/ORCID/S2/GS/PubPeer/RW/arXiv 自动抓取 | config.yaml (scholar name + IDs) | auto_fetched.json | Step 2 | international | init 完成 |
| openalex_enricher | scripts/domestic/openalex_enricher.py | 用 OpenAlex 补充国内论文的国际引用数据 | 论文 JSON | 增强后 JSON | Step 2 后 | domestic | data_importer |
| xiaohongshu_client | scripts/international/xiaohongshu_client.py | 小红书学生评价抓取 + 情感/维度提取 | 导师名/学校名 | 评价 JSON | Step 6 | international/cross_border | — |
| wechat_search | scripts/domestic/wechat_search.py | 微信公众号文章搜索（补充线索） | 关键词 | 文章列表 | Step 6 | domestic | — |
| review_matcher | scripts/domestic/review_matcher.py | 研学网评价表匹配 → investigation_leads | 导师名/学校 | 结构化 leads JSON | Step 6 | domestic | _private/研学网导师评价表.xlsx |
Analysis（分析）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| text_profiler | scripts/analysis/text_profiler.py | PDF/Markdown/文本分析: 词频、原创性标记、引用模式、论文类型分类 | PDF/Markdown 文件 | JSON profile | Step 3 | all | 论文 PDF 获取 |
| paper_quality_rubric | scripts/analysis/paper_quality_rubric.py | 论文六维评分（创新性/方法/论证/文献/写作/贡献） | text_profiler 输出 JSON | 六维评分 JSON | Step 3 | all | text_profiler |
| hybrid_scorer | scripts/analysis/hybrid_scorer.py | 综合评分（合并多维度） | 多个评分 JSON | 综合分 JSON | Step 3 | all | paper_quality_rubric |
| stylometry_profiler | scripts/analysis/stylometry_profiler.py | 风格计量学：虚词频率、句法结构、相似度矩阵 | 多篇文本 | 相似度热力图 + JSON | Step 5 代笔检测 | all | 多篇 PDF |
| citation_profiler | scripts/analysis/citation_profiler.py | 引用分析：自引率、互引、引用质量分布 | 论文 JSON (含引用) | 引用分析 JSON | Step 4/5 | all | data_importer/data_fetcher |
| citation_verifier | scripts/analysis/citation_verifier/ | AI 引用核查：从 .docx/.tex 提取引用，两级 LLM 验证（存在性/元数据/语义支持度） | .docx 或 .tex+.bib 文件 | 引用核查 JSON+CSV 报告 | Step 5 | all | LLM API key |
| common_heuristics | scripts/analysis/common_heuristics.py | 共享异常检测规则 (C01-C07): 产出量、期刊集中度、引用模式等 | 论文列表 + author_profile | 异常 flags 列表 | Step 5 | all | Step 2 数据 |
| review_aggregator | scripts/analysis/review_aggregator.py | 多源评价合并（研学网 + 小红书 + RMP） | 多个评价 JSON | 合并评价 JSON | Step 6 | all | 各评价源 |
| journal_credibility_checker | scripts/analysis/journal_credibility_checker.py | 期刊可信度评估（掠夺性期刊检测） | ISSN/期刊名列表 | 风险评级 JSON | Step 5 | all | — |
| source_evaluation | scripts/analysis/source_evaluation.py | CRAAP Test 信息源评估（时效性/相关性/权威性/准确性/目的性） | 信息源 URL/描述 + 五维评分 | 评估报告 JSON | Step 6 | all | — |
Network（关系网络）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| network_visualizer | scripts/network/network_visualizer.py | D3.js 交互式关系网络图 | relationship_network JSON | {name}_network.html | Step 4 | all | Step 2 数据 |
| timeline_weaver | scripts/network/timeline_weaver.py | 时间线编织（事件序列可视化） | career_timeline JSON | 时间线 HTML | Step 4 | all | Step 1 |
| grant_linker | scripts/network/grant_linker.py | 基金项目关联分析 | 基金数据 JSON | 关联图 JSON | Step 4 | domestic | — |
| negative_space_analyzer | scripts/network/negative_space_analyzer.py | 负面空间分析（官方通报回避了什么） | 通报文本 | evasion_score matrix | Step 5 | all | — |
| citation_constellation | scripts/network/citation_constellation.py | 引用星座图（引用关系可视化） | 引用数据 JSON | 星座图 HTML | Step 4/5 | all | citation_profiler |
| investigation_retrospector | scripts/network/investigation_retrospector.py | 调查复盘（提取可复用签名） | 完成的案件数据 | heuristics 更新建议 | 案件结束后 | all | — |
Deep Evidence / Data Forensics（数据取证）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| data_integrity_checker | scripts/deep_evidence/data_forensics/data_integrity_checker.py | 数据造假统计指纹检测（尾数/小数位/重复） | Excel/CSV 原始数据 | risk_score + findings JSON | Step 5 有原始数据时 | all | 人工提取表格数据 |
| stats_reverse_engineer | scripts/deep_evidence/data_forensics/stats_reverse_engineer.py | 统计反推一致性检验（均值/SD/n → t/F 值验证） | 论文表格统计量 | 不一致性标记 JSON | Step 5 | all | 人工提取统计量 |
| image_metadata_extractor | scripts/deep_evidence/data_forensics/image_metadata_extractor.py | 图像元数据提取（EXIF/创建时间/软件） | 图片文件 | 元数据 JSON | Step 5 | all | 论文图片提取 |
Deep Evidence / Publication Trace（发表链追踪）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| preprint_monitor | scripts/deep_evidence/publication_trace/preprint_monitor.py | 预印本监控（arXiv/bioRxiv/SSRN） | 作者名/DOI | 预印本-期刊映射 JSON | Step 5 | all | — |
| conference_paper_mapper | scripts/deep_evidence/publication_trace/conference_paper_mapper.py | 会议论文→期刊论文转化追踪 | 论文列表 JSON | 转化映射 JSON | Step 5 | all | Step 2 数据 |
| bilingual_publication_detector | scripts/deep_evidence/publication_trace/bilingual_publication_detector.py | 双语发表检测（中英文重复发表） | 合并论文列表 | 疑似重复对 JSON | Step 5 | cross_border | merger 输出 |
| crossref_event_tracker | scripts/deep_evidence/publication_trace/crossref_event_tracker.py | Crossref 事件追踪（更正/撤稿/表达关注） | DOI 列表 | 事件 JSON | Step 5 | all | Step 2 数据 |
Deep Evidence / Ethics & Peer Review（伦理与同行评议）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| ethics_statement_parser | scripts/deep_evidence/ethics_audit/ethics_statement_parser.py | 伦理声明解析（IRB 批号提取与验证） | 论文 PDF/文本 | 伦理声明结构化 JSON | Step 5 | all | 论文 PDF |
| clinical_trial_registry_checker | scripts/deep_evidence/ethics_audit/clinical_trial_registry_checker.py | 临床试验注册核查（ClinicalTrials.gov/ChiCTR） | 注册号列表 | 注册状态 JSON | Step 5 | all | ethics_statement_parser |
| review_cycle_analyzer | scripts/deep_evidence/peer_review_intel/review_cycle_analyzer.py | 审稿周期分析（投稿→接收时间异常检测） | 论文元数据 JSON | 周期异常 flags | Step 5 | all | Step 2 数据 |
| editorial_self_publishing_detector | scripts/deep_evidence/peer_review_intel/editorial_self_publishing_detector.py | 编委自发文检测（编委任期内发文模式） | 编委名单 + 论文列表 | 自发文率 JSON | Step 5 | all | 人工收集编委名单 |
| recommended_reviewer_network | scripts/deep_evidence/peer_review_intel/recommended_reviewer_network.py | 推荐审稿人网络分析 | 审稿人数据 | 网络图 JSON | Step 5 | all | — |
| journal_retraction_history | scripts/deep_evidence/peer_review_intel/journal_retraction_history.py | 期刊撤稿历史查询 | 期刊名/ISSN | 撤稿记录 JSON | Step 5 | all | — |
Show full SKILL.md (502 more words)Show less
Deep Evidence / Evidence Compiler（证据链编译）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| signal_aggregator | scripts/deep_evidence/evidence_compiler/signal_aggregator.py | 多模块信号聚合（统一格式汇总所有异常信号） | 各模块输出 JSON | 聚合信号 JSON | Step 5 末尾 | all | 各检测模块 |
| evidence_chain_builder | scripts/deep_evidence/evidence_compiler/evidence_chain_builder.py | 证据链构建（信号→假设→证据链） | 聚合信号 JSON | 证据链 JSON + 可视化 | Step 7 前 | all | signal_aggregator |
Benchmark & Matching（基准线与匹配）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| benchmark_engine | scripts/benchmark_engine.py | 学科基准线数据库：5层异常评分（Z-score/对数正态/t分布） | 结构化指标 (h-index, 年均论文等) | composite_score + risk_level JSON | Step 5 | all | Step 2 数据 |
| scholar_profile_matcher_v2 | scripts/scholar_profile_matcher_v2.py | 17维特征向量匹配：与46案例库对比 + 不端模式相似度 | scholar_data JSON | 匹配报告 JSON | Step 5 后 | all | Step 3 + Step 5 |
Validation（校验）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| data_validator (domestic) | scripts/domestic/data_validator.py | 国内 scholar_data JSON schema + 逻辑校验 | scholar_data JSON | 校验报告 | Step 7 前 | domestic | scholar_data_builder |
| data_validator (international) | scripts/international/data_validator.py | 国际 scholar_data schema + 逻辑校验 | scholar_data JSON | 校验报告 | Step 8 前 | international | scholar_data_builder |
| cross_border validator | scripts/cross_border/validator.py | 跨境一致性检查（时间线重叠、学历真伪） | merged scholar_data | 一致性报告 JSON | Step 3 (cross_border) | cross_border | merger |
| scholar_data_builder (domestic) | scripts/domestic/scholar_data_builder.py | 构建统一 scholar_data.json | config + 各脚本输出 | scholar_data.json | 各步骤输出后 | domestic | — |
| scholar_data_builder (international) | scripts/international/scholar_data_builder.py | 构建国际 scholar_data.json | config + auto_fetched + xhs | scholar_data.json | 各步骤输出后 | international | — |
| cross_border merger | scripts/cross_border/merger.py | 合并国内+国际 scholar_data | 两套 JSON | merged JSON | Step 2 后 | cross_border | 两个 builder |
Report & Delivery（报告与交付）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track | 依赖 |
|---|---|---|---|---|---|---|---|
| report_prompt_optimizer | scripts/report/report_prompt_optimizer.py | 报告 prompt 优化（注入语言规范 + 置信度体系） | 报告草稿 | 优化后 prompt | Step 7 | all | — |
| md_to_pdf | scripts/md_to_pdf.py | Markdown→PDF（A4/封面/TOC/图表自动生成） | Markdown 报告 | styled PDF | Step 7 后 | all | chart_generator |
| chart_generator | scripts/chart_generator.py | 自动生成图表（雷达/饼图/热力图/时间线/网络图） | Markdown 中 <!--chart:--> 注解 | PNG 图表 | Step 7 | all | — |
Orchestration（编排）
| 工具 | 路径 | 功能 | 输入 | 输出 | 时机 | Track |
|---|---|---|---|---|---|---|
| investigate.py | scripts/investigate.py | CLI 主入口：init/import/generate/smart-step 等子命令 | CLI args + config.yaml | 各步骤输出 | 全程 | all |
| investigate_visual.py | scripts/investigate_visual.py | Rich 终端可视化包装 + smart-step 自动推进 | 同 investigate.py | 彩色终端输出 | 全程 | all |
三 Track 路由表
| Track | 判定条件 | 入口命令 | 详细流程文档 |
|---|---|---|---|
| domestic | 学者仅在国内机构任职，无海外学位/教职 | investigate.py init --type domestic | docs/skill/workflow_domestic.md |
| international | 外国学者/导师，或仅在海外机构任职 | investigate.py init --type international | docs/skill/workflow_international.md |
| cross_border | 学者同时具有国内任职 + 海外学位/教职经历 | investigate.py init --type cross_border | docs/skill/workflow_cross_border.md |
7 步框架概览
| Step | 名称 | 核心动作 | 主要工具 |
|---|---|---|---|
| 0 | Case Registry | 案件注册，生成唯一 ID | case_manager.py |
| 1 | Basic Profile | 建立身份时间线 | config.yaml + investigate.py init |
| 2 | Output Quantity | 核实声称 vs 实际产出 | data_importer / data_fetcher |
| 3 | Quality Assessment | 论文质量六维评分 | text_profiler → paper_quality_rubric |
| 4 | Relationship Network | 合作关系与资源依赖图谱 | network_visualizer + grant_linker |
| 5 | Anomaly Detection | 异常检测 + 深度证据 | common_heuristics + deep_evidence/* + benchmark_engine |
| 6 | Multi-Source Validation | 多源交叉验证 | review_aggregator + wechat_search / xiaohongshu_client |
| 7 | Report Generation | 报告生成 + 交付 | report_prompt_optimizer → md_to_pdf → watermark |
📖 每步的详细流程、证据标准、红旗信号，请 Read
docs/skill/workflow_domestic.md（国内）或对应 track 文档。
引用核查补充流程（LLM 操作）
当需要验证论文中引用的真实性时，LLM 按以下三步操作。
📖 判定框架（造假类型学 A–E）、判据顺序（先全网题名检索、再期次目录穷尽）、验证源优先级与字段级核对表模板，请先 Read
docs/skill/citation_fabrication_audit.md。核查必须 100% 逐条覆盖，禁止抽样。
Step 1: 提取引用
python scripts/citation_verifier.py -i paper.docx -o output/citation_task.json --extract-only输出一个 JSON 文件，每条引用包含 key、ref_text（参考文献原文）、context_block（引用在正文中的使用上下文），以及空的 verification 字段。
⚠️ 提取盲区：docx 解析不覆盖脚注/尾注。脚注或尾注制稿件须另行抽取
word/footnotes.xml与word/endnotes.xml（命令见docs/skill/citation_fabrication_audit.md的 Extraction Caveats 节），否则引注会整片漏掉。
Step 2: LLM 逐条验证
读取 Step 1 输出的 JSON，对每条引用执行以下核查：
| 检查项 | 方法 | 填入字段 |
|---|---|---|
| 引用是否存在 | 搜索 Google Scholar / 查 DOI / 查期刊官网 | citation_exists |
| 元数据是否正确 | 对比搜索结果中的作者、年份、标题 | citation_integrity + citation_integrity_details |
| 是否支持文内断言 | 阅读摘要或全文，判断引用是否支撑了论文中的说法 | citation_usage_correct + citation_usage_reasoning |
| 处理建议 | 综合判断保留/替换/移除 | recommendation + suggested_replacement |
多引用规则：当正文用 (1, 2) 同时引用多篇文献时，只要被核查的这篇支持断言的某一部分即可标记为 Yes。
填写完成后保存 JSON。
Step 3: 生成报告
python scripts/citation_verifier.py --report output/citation_task.json -o output/citation_report.json输出结构化 JSON 报告（含统计摘要 + 问题引用列表）和 CSV 快速审阅文件。
何时使用
- Step 3 质量评估中发现引用可疑时
- Step 5 异常检测中需要核实引用真实性时
- 用户直接提供论文 PDF 要求审查引用时
PDF 直接审查补充流程
当用户直接提供论文 PDF（而非结构化数据）时：
- Read PDF → 提取文本、表格、Figure caption
- 六式扫描 → 图片复用/数据造假/图片拼接/统计异常/产出异常/方法矛盾
- 数值提取 → 从表格提取数值，保存 CSV → 可选喂给
data_integrity_checker.py - 交叉验证 → 多个可疑点是否指向系统性造假
- 报告输出 → 结构化 Markdown 审查报告
⚠️ LLM 对图片的分析基于视觉理解，无法进行像素级 ELA。需精确图像比对时标注"建议使用 ImageTwin/Forensically 进一步验证"。
分层加载指令
执行具体步骤时，按需 Read 以下文档：
| 场景 | Read 文档 |
|---|---|
| 执行国内 track 某步骤的详细流程 | docs/skill/workflow_domestic.md |
| 执行国际 track 某步骤的详细流程 | docs/skill/workflow_international.md |
| 执行海归 track 的合并/验证流程 | docs/skill/workflow_cross_border.md |
| 需要了解 13+8 种调查类型定义 | docs/skill/investigation_types.md |
| 核查参考文献真实性（造假类型学/判据顺序/字段核对表） | docs/skill/citation_fabrication_audit.md |
| 需要扩展调查模块 A-E 的详细方法 | docs/skill/extended_modules.md |
| 需要深度证据层的架构设计细节 | docs/skill/deep_evidence_layer.md |
| 需要多智能体/交付层的协作协议 | docs/skill/multi_agent_delivery.md |
| 需要导师蒸馏服务的 API 接口 | docs/skill/mentor_distill.md |
| 生成报告前的交付物检查清单 | docs/skill/report_checklist.md |
| 需要启发式规则库（结构签名 S1-S10） | scripts/heuristics.md |
| 需要报告语言规范（置信度 L1-L5） | scripts/report_language_spec.md |
© yadnuses, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 344 other files (scripts) in the repository root of yadnuses/Academic-Detective.
- SKILL.md
- .github/ISSUE_TEMPLATE/apply_for_investigation.md
- .github/ISSUE_TEMPLATE/claim_investigation.md
- .github/PULL_REQUEST_TEMPLATE/contribute_case.md
- .github/workflows/verify_contribution.yml
- .gitignore
- CODE_OF_CONDUCT.md
- CONTRIBUTING.md
- DESCRIPTION.txt
- LICENSE
- README.md
- _private/.gitkeep
- archive/SKILL_v2_full.md
- archive/flat_export_redundant_20260501/README.md
- … and 331 more
Open the folder on GitHubat commit 8de7b96
Compare with similar skills
Academic Investigation next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Academic Investigation this skillyadnuses/Academic-Detective | 313 | — | ~4.2k | Automated safety check: Pass | MIT | |
| MathModel Figure Templatesjihe520/MathModelAgent | 6.2k | — | ~627 | Automated safety check: Notes | None | |
| Scientific Figure MakingChenLiu-1996/figures4papers | 8.3k | — | ~557 | Automated safety check: Pass | Custom licence | |
| Academic Figure SkillTingxiYu/academic-figure-skill | 483 | 1 repos | ~7k | Automated safety check: Pass | Apache-2.0 | |
| FigMirror Figure Style TransferVILA-Lab/FigMirror | 521 | — | ~2.1k | Automated safety check: Pass | None | |
| FigMirrorVILA-Lab/FigMirror | 521 | — | ~2.4k | Automated safety check: Pass | None |
Similar skills
Scientific Figure Making
ChenLiu-1996/figures4papers
Covers publication-ready matplotlib figures for academic papers, slides, and reports—bars, trends, scatter, heatmaps, and multi-panel layouts—with this…
Academic Figure Skill
TingxiYu/academic-figure-skill
Academic-grade scientific figure creation for Nature/Cell/Science journals.
FigMirror Figure Style Transfer
VILA-Lab/FigMirror
Redraws your data as a matplotlib figure in the visual style of a reference paper figure, using a drawer and reviewer loop.
FigMirror
VILA-Lab/FigMirror
Mirrors the visual style of a top-conference paper figure onto your own data, producing a camera-ready PDF and a self-contained matplotlib script.
ModelViz Scientific Plots
hrdZhu/modelviz-skill
Turns your CSV or Excel data and a plain-language request into a publication-style scientific chart by adapting a catalog template, then checks and repairs it.
Related
Categories
Questions about Academic Investigation
What does Academic Investigation do?
Academic integrity investigation and scholar profile analysis system. Academic Investigation is an agent skill from yadnuses/Academic-Detective. Academic integrity investigation and scholar profile analysis system.
When should I use Academic Investigation?
Academic Investigation fits situations like: conducting comprehensive academic background checks; credential verification; publication analysis; institutional affiliation audits of scholars.
How do I install Academic Investigation in Claude Code?
Run `npx skills add yadnuses/Academic-Detective --skill academic-investigation -a claude-code`. Or copy the skill folder (the yadnuses/Academic-Detective repository) into .claude/skills/academic-investigation in your project. Claude Code loads it when a task matches its description.
How do I install Academic Investigation in Codex?
Run `npx skills add yadnuses/Academic-Detective --skill academic-investigation -a codex`. Or copy the skill folder (the yadnuses/Academic-Detective repository) into .agents/skills/academic-investigation in your project. Codex loads it when a task matches its description.
Can I use Academic Investigation in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add yadnuses/Academic-Detective --skill academic-investigation -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/academic-investigation, .gemini/skills/academic-investigation, .github/skills/academic-investigation and .opencode/skills/academic-investigation in your project.
What does Academic Investigation need to run?
Going by SKILL.md and its folder, Academic Investigation needs the command-line tools its instructions call (python). Our summary lists: Python 3.
Does Academic Investigation access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Academic Investigation safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Academic Investigation use?
Academic Investigation is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Academic Investigation use?
About 4.2k tokens (SKILL.md is roughly 17k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Academic Investigation?
Skills that share tags, products or a category with Academic Investigation: MathModel Figure Templates (jihe520/MathModelAgent, 6.2k stars), Scientific Figure Making (ChenLiu-1996/figures4papers, 8.3k stars), Academic Figure Skill (TingxiYu/academic-figure-skill, 483 stars) and FigMirror Figure Style Transfer (VILA-Lab/FigMirror, 521 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Academic Investigation?
yadnuses (a GitHub user) maintains it in yadnuses/Academic-Detective, which has 313 GitHub stars. The repository was last updated on October 10, 2026.
Source: yadnuses/Academic-Detective on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.