Role
Best agent skills for researchers, page 132
Skills for researchers, ranked
Ranked by score. Sort bymost stars,trending,newest,recently updated
| # | Skill | Repository | Stars | Used in | Tokens | Auto-check | Licence | Updated |
|---|---|---|---|---|---|---|---|---|
| 6289 | 6289.Extract Builds the gauntlet knowledge base from AST extraction and AI enrichment. | athola/ | 341 | — | ~510 | Automated safety check: Pass | MIT | 2 days ago |
| 6290 | 6290.Graph Build Builds or updates the code knowledge graph via tree-sitter AST and SQLite. | athola/ | 341 | — | ~609 | Automated safety check: Pass | MIT | 2 days ago |
| 6291 | 6291.Graph Search Searches the code knowledge graph by function, class, or type using FTS5 full-text search. | athola/ | 341 | — | ~555 | Automated safety check: Pass | MIT | 2 days ago |
| 6292 | 6292.Research Runs multi-source research across GitHub, HN, Reddit, arXiv, and Semantic Scholar. | athola/ | 341 | — | ~2.3k | Automated safety check: Pass | MIT | 2 days ago |
| 6293 | 6293.Msa Gauge Rr Measurement System Analysis (MSA) and Gauge Repeatability & Reproducibility (Gauge R&R) — plan, execute, and interpret an MSA study for variable or attribute measurement systems. | hashgraph-online/ | 1.3k | — | ~2.5k | Automated safety check: Pass | MIT | yesterday |
| 6294 | Write and revise rigorous research narratives, literature reviews, manuscript sections, citations, and figure discussions. | hashgraph-online/ | 1.3k | — | ~1.1k | Automated safety check: Pass | Apache-2.0 | yesterday |
| 6295 | 6295.Customer Research Research customer questions by searching across documentation, knowledge bases, and connected sources, then synthesize a confidence-scored answer. | w95/ | 244 | — | ~2.3k | Automated safety check: Pass | MIT | 7 mo ago |
| 6296 | 6296.Knowledge Base You are about to research a topic, answer a factual question about the org, or look up context about a person, project, or tool. | grandamenium/ | 101 | — | ~673 | Automated safety check: Pass | MIT | 19 days ago |
| 6297 | 6297.Research Log Maintain research logs and decision records for academic projects. | Drchronx/ | 139 | — | ~299 | Automated safety check: Pass | Unknown | 4 mo ago |
| 6298 | Track journal submission, reviewer comments, revisions, and response letters. | Drchronx/ | 139 | — | ~365 | Automated safety check: Pass | Unknown | 4 mo ago |
| 6299 | Curated tools and resources for effective scientific writing | brycewang-stanford/ | 4.6k | — | ~1.9k | Automated safety check: Pass | Unknown | 6 days ago |
| 6300 | 6300.Submit Submission pipeline — journal targeting, replication package, audit, and final gate. | brycewang-stanford/ | 4.6k | — | ~591 | Automated safety check: Notes | Unknown | 6 days ago |
| 6301 | 6301.Biomedical Skills 24 biomedical research skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~827 | Automated safety check: Pass | MIT | 3 mo ago |
| 6302 | 6302.Citation Skills 22 citation management skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~812 | Automated safety check: Pass | MIT | 3 mo ago |
| 6303 | 6303.Deep Research Skills 13 deep research & systematic reviews skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~535 | Automated safety check: Pass | MIT | 3 mo ago |
| 6304 | 6304.Fulltext Skills 16 full-text access skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~577 | Automated safety check: Pass | MIT | 3 mo ago |
| 6305 | 6305.Metadata Skills 24 metadata & bibliometrics skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~836 | Automated safety check: Pass | MIT | 3 mo ago |
| 6306 | 6306.Methodology Skills 12 research methodology skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~496 | Automated safety check: Pass | MIT | 3 mo ago |
| 6307 | 6307.Search Skills 31 database search skills. An agent skill from wentorai/research-plugins. | wentorai/ | 298 | 1 repo | ~1.1k | Automated safety check: Pass | MIT | 3 mo ago |
| 6308 | CD-GPT 生成式生物基础大模型,对 DNA/RNA/蛋白质序列预训练,支持单分子与多分子联合分析的预测与生成. An agent skill from infometa/workbuddyskills. | infometa/ | 348 | — | ~5k | Automated safety check: Pass | No licence | 2 days ago |
| 6309 | 6309.Kidd Script Writing Creates deep-analysis short video scripts in "吟游诗人基德" (Minstrel Kidd) style. | infometa/ | 348 | — | ~1.3k | Automated safety check: Pass | No licence | 2 days ago |
| 6310 | 乐享知识库 MCP 全功能 Skill。当用户提到「乐享」「知识库」「个人知识库」「我的知识库」「lexiang」,或提供 lexiangla.com 链接,或涉及知识库的搜索/写入/编辑/文件/配置/评论/草稿/智能表格等操作时使用。 | infometa/ | 348 | — | ~1.2k | Automated safety check: Pass | No licence | 2 days ago |
| 6311 | This skill should be used when the user is NOT a parent seeking counseling, but instead a researcher, journalist, editor, public speaker, policy maker, content creator, or training instructor who… | infometa/ | 348 | — | ~1.3k | Automated safety check: Pass | No licence | 2 days ago |
| 6312 | 调用肺炎AI辅助诊断模型API,支持提交胸部CT DICOM压缩包进行异步分析、查询分析结果、以及自动轮询等待结果完成。采用HMAC-SHA256签名鉴权。触发词:肺炎检测、肺炎AI、胸部CT分析、pneumonia、DICOM、肺炎查询。 | infometa/ | 348 | — | ~1.9k | Automated safety check: Pass | No licence | 2 days ago |
| 6313 | 微信公众号文章检索工具。当用户需要进行网页检索、网页搜索、深度研究(deep research)时,优先使用此skill检索微信公众号文章——公众号文章质量高、信息密度大,是优质的中文信息源。基于搜狗微信搜索接口实现。 | infometa/ | 348 | — | ~479 | Automated safety check: Pass | No licence | 2 days ago |
| 6314 | 6314.Methods Reporting Check methods reporting against CONSORT, JARS, DA-RT standards. | brycewang-stanford/ | 4.6k | — | ~4.7k | Automated safety check: Pass | Unknown | 6 days ago |
| 6315 | End-to-end protein design pipeline guide across preparation, generation, validation, and filtering. | BioTender-max/ | 200 | — | ~1.4k | Automated safety check: Pass | MIT | 3 mo ago |
| 6316 | Run a multi-agent scientific manuscript review with parallel specialist reviewers, disagreement checks, and an editor meta-review. | BioTender-max/ | 200 | — | ~2.2k | Automated safety check: Pass | Unknown | 3 mo ago |
| 6317 | 6317.Protein Design Qc Protein design quality control, filtering thresholds, and ranking guidance. | BioTender-max/ | 200 | — | ~2.6k | Automated safety check: Pass | MIT | 3 mo ago |
| 6318 | 6318.Proteina Complexa Proteina-Complexa flow-based protein backbone generation with fold-conditioned sampling guidance. | BioTender-max/ | 200 | — | ~1.4k | Automated safety check: Notes | MIT | 3 mo ago |
| 6319 | Chai-1 structure prediction for protein complexes and design validation. | BioTender-max/ | 200 | — | ~1.7k | Automated safety check: Pass | MIT | 3 mo ago |
| 6320 | 6320.Distill Memory Capture breakthrough moments and valuable insights as searchable memories in your knowledge base. | nowledge-co/ | 186 | — | ~486 | Automated safety check: Pass | No licence | yesterday |
| 6321 | 6321.Deep Research Conduct in-depth, multi-step research on a given topic by decomposing queries, finding diverse sources, cross-referencing findings, and synthesizing a comprehensive report. | seb1n/ | 206 | — | ~2.2k | Automated safety check: Pass | MIT | 2 mo ago |
| 6322 | 6322.Fact Checking Verify the accuracy of claims and statements by extracting individual assertions, identifying authoritative sources, cross-referencing evidence, and assigning confidence-scored verdicts. | seb1n/ | 206 | — | ~2.4k | Automated safety check: Pass | MIT | 2 mo ago |
| 6323 | Build structured knowledge graphs from unstructured text by extracting entities, mapping relationships, generating graph triples, and visualizing the result. | seb1n/ | 206 | — | ~2.5k | Automated safety check: Pass | MIT | 2 mo ago |
| 6324 | 6324.Note Taking Capture, organize, and retrieve notes efficiently using structured formats, tagging, and file management for meetings, ideas, research, and daily logs. | seb1n/ | 206 | — | ~2.4k | Automated safety check: Pass | MIT | 2 mo ago |
| 6325 | 6325.Literature Review Conduct a structured literature review on a given topic by defining a search strategy, applying inclusion and exclusion criteria, extracting key findings, and synthesizing results into a coherent… | seb1n/ | 206 | — | ~2.6k | Automated safety check: Pass | MIT | 2 mo ago |
| 6326 | This skill should be used when the user asks about "KOL", "citation velocity", "who are the key opinion leaders", "is the science gaining momentum", "publication trend for [target]"… | hh-health-AI/ | 101 | — | ~725 | Automated safety check: Pass | MIT | 3 days ago |
| 6327 | Researcher for converting unit tests to browser tests for Project Bedrock. | nwjs/ | 160 | — | ~433 | Automated safety check: Pass | BSD-3-Clause | yesterday |
| 6328 | 6328.Md Prepare ASE molecular-dynamics workflow tasks with backend-agnostic controls. | jinzhezenggroup/ | 148 | — | ~269 | Automated safety check: Pass | MIT | 2 days ago |
| 6329 | 6329.Md Prepare ABINIT molecular-dynamics task inputs from a user-provided structure and MD controls. | jinzhezenggroup/ | 148 | — | ~296 | Automated safety check: Pass | MIT | 2 days ago |
| 6330 | 6330.Md Prepare CP2K molecular-dynamics task inputs from a user-provided structure and MD controls. | jinzhezenggroup/ | 148 | — | ~297 | Automated safety check: Pass | MIT | 2 days ago |
| 6331 | 6331.Md Prepare SIESTA molecular-dynamics task inputs from a user-provided structure and MD controls. | jinzhezenggroup/ | 148 | — | ~300 | Automated safety check: Pass | MIT | 2 days ago |
| 6332 | 6332.Md Prepare DFTB+ molecular-dynamics task inputs from a user-provided structure and MD controls. | jinzhezenggroup/ | 148 | — | ~291 | Automated safety check: Pass | LGPL-3.0-or-later | 2 days ago |
| 6333 | 6333.Call Fraud Shield Post-call fraud detection skill. An agent skill from CALLE-AI/awesome-phone-call-agents. | CALLE-AI/ | 107 | — | ~2.7k | Automated safety check: Pass | MIT | yesterday |
| 6334 | Post-call cooperation skill. An agent skill from CALLE-AI/awesome-phone-call-agents. | CALLE-AI/ | 107 | — | ~1.1k | Automated safety check: Pass | MIT | yesterday |
| 6335 | 6335.Phone Claim Verifier Verify claims about offline records over the phone with CALL-E - "call these providers and check they still accept this plan", "confirm these supplier contacts are current", "phone-verify this… | CALLE-AI/ | 107 | — | ~1.8k | Automated safety check: Pass | MIT | yesterday |
| 6336 | The full pipeline for idea generation. An agent skill from GRIND-Lab-Core/night_owl_research_agent. | GRIND-Lab-Core/ | 106 | — | ~4.4k | Automated safety check: Warn | No licence | 5 mo ago |
Explore related skills
Categories for researchers
Popular topics
- Bioinformatics1,146
- Citation management933
- Slides and decks807
- Web scraping776
- PDF773
- Translation699
- Markdown627
- Excel spreadsheets626
- Data analysis562
- Statistics541
- CSV and tabular files530
- Brand voice and tone512
- Social media posts509
- Academic paper search507
- Literature review482
- PowerPoint presentations442
- Data pipelines and ETL419
- Plain language and style rules419
- Data visualization417
- LaTeX414
- Deep research407
- Word documents402
- Reproducible research376
- Knowledge bases375
Products these skills work with
Other roles
- Developers39,105
- Frontend Developers7,075
- DevOps and SRE Engineers6,607
- Data Scientists5,833
- Designers4,873
- Marketers5,420
- SEO Specialists1,401
- Writers and Editors5,851
- Product Managers2,029
- Founders and Solopreneurs6,359
- Students and Educators1,988
- Lawyers and Compliance Teams1,330
- Accountants and Finance Teams2,470