brycewang-stanford/awesome-agent-skills-for-empirical-research
96 skills823 installs312k starsGitHub
Install
npx skills add https://github.com/brycewang-stanford/awesome-agent-skills-for-empirical-researchSkills in this repo
1Chinese De Aigcchinese-de-aigc is a skill that reduces AI-detection rates in Chinese academic papers by rewriting structural AI-writing patterns. A researcher uses it to revise machine-drafted Chinese text so it reads like a real author while preserving academic rigor. It applies 17 diagnosed pattern classes through a five-step locate, diagnose, rewrite, self-score, and recheck workflow, with section-specific strategies.72installs2Stata Accounting ResearchA skill providing Stata syntax patterns drawn from 126 peer-reviewed Journal of Accounting Research replication files spanning 2017-2025. Accounting researchers use it to look up how published papers implemented methods like difference-in-differences, event studies, and entropy balancing, then adapt the code to their own variable names. It provides code patterns with source citations but not research-design or identification advice.52installs3Causal Inference MixtapeCausal Inference: The Mixtape is a skill that provides code templates for causal inference methods in Python, R, and Stata. A researcher uses it to implement DiD, event studies, RDD, instrumental variables, synthetic control, and propensity score matching. It is built from Scott Cunningham's Causal Inference: The Mixtape and includes cross-language equivalents and robustness-check patterns.47installs4Python Econ ComputingA skill that guides an agent to write Python code for macroeconomic modeling and quantitative economic analysis. It covers DSGE and HANK models, numerical methods, and causal inference, with a library reference mapping each task to tools like numpy, numba, quantecon, sequence_jacobian, pyfixest, and econml. A researcher uses it when building economic models or running causal designs in Python. Its core rule is to vectorize first, accelerate loops with Numba, and keep code aligned with economic theory.43installs5MarginaleffectsThis skill is a manual for the marginaleffects R and Python package and its companion book Model to Meaning. A data scientist uses it to compute predictions, comparisons, and slopes, estimate treatment effects like ATE/ATT/CATE, test hypotheses, and do G-computation for causal inference. It organizes each analysis around a five-question framework covering estimand, grid, aggregation, uncertainty, and testing.40installs6Literature ReviewThis skill conducts systematic, comprehensive literature reviews across multiple academic databases such as PubMed, arXiv, bioRxiv, and Semantic Scholar. A researcher uses it to plan a search with the PICO framework, search databases, synthesize findings thematically, verify citations, and generate professional markdown and PDF documents. It targets biomedical, scientific, and technical domains and requires at least one generated figure such as a PRISMA flow diagram.39installs7Avoid Ai WritingAvoid-ai-writing audits and rewrites content to remove AI writing patterns called AI-isms. A writer uses it to clean up text that sounds machine-generated, with a rewrite mode that fixes issues and a detect-only mode that flags them. It targets formatting problems like em dashes and bold overuse, sentence structures like 'not X, it's Y', and a tiered list of overused words. It works with any AI coding assistant that supports the SKILL.md format.38installs8Full Empirical Analysis SkillThis skill runs a full 8-step empirical analysis pipeline in Python using pandas, statsmodels, linearmodels, pyfixest, econml, and causalml. An applied economist uses it to go from data cleaning through descriptive statistics, diagnostics, baseline models, robustness checks, and publication-ready AER-style tables and figures. It adds parallel modes for epidemiology and ML causal inference.38installs9Full Empirical Analysis Skill StataThis skill runs a full 8-step empirical analysis pipeline in Stata using reghdfe, ivreg2, csdid, rdrobust, psmatch2, and esttab. An applied economist uses it for a reproducible .do-file workflow from data cleaning through robustness checks and publication-ready tables and figures. It adds parallel modes for epidemiology and ML causal inference.38installs10Latex DocumentThis skill creates, compiles, and converts LaTeX documents into professional PDFs with PNG previews. A user reaches for it to build resumes, reports, theses, presentations, posters, exams, invoices, or fillable forms, and to convert between Markdown, DOCX, HTML, and LaTeX. It also OCRs handwritten or printed PDFs into LaTeX and supports pdflatex, xelatex, and lualatex.37installs11Stop SlopA prose-editing skill that removes predictable AI writing patterns from text. Writers use it when drafting, editing, or reviewing content to eliminate filler phrases, formulaic structures, passive voice, and vague declaratives. It provides eight core rules, a quick pre-delivery checklist, and a five-dimension scoring rubric with a revise threshold.36installs12DeslopDeslop is a Claude skill that removes AI writing patterns from prose so text reads like a specific human wrote it. A writer uses it while drafting or editing manuscripts, abstracts, grant narratives, blog posts, or reports to cut filler, break formulaic structures, and eliminate AI tropes. It ships reference catalogs of phrases, structures, and tropes plus a pre-delivery checklist.35installs13Humanizer AcademicThis skill removes signs of AI-generated writing from academic medical manuscripts. An author or editor uses it to detect and rewrite patterns like inflated significance claims, superficial -ing phrases, vague attributions, and excessive hedging, while keeping the scientific content intact. It is based on Wikipedia's 'Signs of AI writing' guide adapted for medical literature.35installs14Full Empirical Analysis Skill RThis skill runs a full 8-step empirical analysis pipeline in R using dplyr, fixest, did, rdrobust, MatchIt, grf, and modelsummary. An applied economist uses it for a tidyverse-style reproducible R or Quarto workflow from data import through robustness and publication-ready tables and figures. It adds parallel modes for epidemiology and ML causal inference.34installs15Humanize ChineseThis skill detects and rewrites AI-generated Chinese text to sound more human. A writer uses its unified CLI to score text for AI patterns, humanize it with best-of-N rewriting, and apply style transforms such as academic, zhihu, or novel. It runs pure Python with no dependencies and includes academic AIGC reduction for Chinese paper checkers.34installs16Statspai SkillA workflow skill that drives StatsPAI (1,100+ registered functions) through the canonical pipeline of an applied empirical paper. Researchers use it to run a full causal analysis in AER/QJE style, epidemiology style (target-trial emulation, IPTW, TMLE), or ML-causal style (DML, meta-learners, causal forest), with each step emitting a paper-ready artifact. It also exports multi-column regression tables to Word, Excel, and LaTeX.32installs17Test SkillA router skill (named auto-empirical-research-skills) used when the full AERS repository is installed as one skill folder. It classifies an empirical-research request by stage, then loads only the matching vendored child skill for causal inference, econometrics, replication, manuscript writing, citation checking, or de-AIGC editing. It directs the agent to the catalog and taxonomy files instead of reading the whole repository at once.23installs18Ml Paper WritingThis skill guides writing publication-ready ML, AI, and systems papers for top conferences such as NeurIPS, ICML, ICLR, ACL, AAAI, COLM, OSDI, NSDI, ASPLOS, and SOSP. A researcher uses it to draft a paper from a research repo, structure arguments, and verify citations. It stresses fetching citations programmatically rather than from memory, and includes LaTeX templates and conference checklists.6installs19Academic Paper VerifyAcademic-paper-verify audits an academic paper against its source R scripts and output files. A researcher uses it to replicate or verify a study, cross-checking LaTeX tables against R output, validating modeling choices, and confirming sample sizes are consistent. It runs six phases: discovery, table audit, inline claims audit, code review, manifest build, and automated replication. It flags any coefficient, claim, or count that does not match the code.5installs20Anti HallucinationAnti-hallucination is a set of rules that govern how the agent verifies information and avoids fabricating functions, arguments, APIs, file paths, data structures, or facts. It applies to all tasks and instructs the agent to read before writing, run and fix code rather than just generate it, and state uncertainty explicitly. A developer uses it to reduce fabricated outputs in coding and analysis work. The rules override any inclination to guess.5installs21Academic PaperAcademic-paper is a paper-writing skill that runs a 12-agent pipeline covering configuration, literature search, structure design, drafting, citation compliance, peer review, and output formatting. A researcher uses it to write, revise, or plan an academic paper across disciplines. It supports multiple paper structures and citation formats, bilingual abstracts in Traditional Chinese and English, and multi-format output. A plan mode guides first-time writers chapter by chapter.4installs22Academic Paper ReviewerAcademic-paper-reviewer simulates a journal peer review with five dynamically configured reviewers. A researcher uses it before submission to get a structured critique of a manuscript from methodology, domain, cross-disciplinary, and core-argument angles. It identifies the paper's field, configures reviewer personas, runs parallel reviews, and synthesizes an editorial decision and revision roadmap. Modes include full review, re-review verification, and a guided learning mode.4installs23Academic PipelineAcademic-pipeline is an orchestrator for the full research-to-publication workflow. A researcher uses it to move from research through writing, integrity checking, review, revision, and finalization in a staged process. It does not do the substantive work itself; it detects the current stage, recommends modes, dispatches the deep-research, academic-paper, and academic-paper-reviewer skills, and tracks state. Mandatory integrity gates require 100 percent reference and data verification before proceeding.4installs24Academic WritingAcademic-writing generates original, publication-ready draft text for academic economics content. A researcher uses it to draft outlines, abstracts, introductions, data and methods sections, results narratives, and referee responses. It follows economics conventions, uses a claim-support-implication paragraph structure, and matches causal language to the study design. It refuses to fabricate estimates, sample sizes, or citations, using labeled placeholders instead.4installs25Deep ResearchDeep Research is a Claude skill that runs a 13-agent team through a full academic research pipeline: question formulation, literature search, source verification, synthesis, and APA 7.0 report compilation. A researcher uses it for literature reviews, systematic reviews with meta-analysis, fact-checking, or Socratic-guided scoping when they have no clear question yet. It offers seven modes ranging from a quick brief to a full PRISMA systematic review.4installs26Econ Abstract WritingEcon Abstract Writing is a Claude skill that guides writing the abstract of an academic economics paper. A researcher uses it to compress findings into four to seven sentences with a results-heavy structure, following patterns from David Evans and Marc Bellemare and top journals like AER and QJE. It advises leading with the research question and method and spending most space on specific results.4installs27Econ Intro WritingEcon Intro Writing is a Claude skill that guides writing the introduction to an academic economics paper. A researcher uses it to structure the intro across motivation, research question, empirical approach, detailed results, and value-added, following analysis from David Evans, Keith Head, Claudia Sahm, and others. It front-loads the paper's contribution and pushes literature discussion toward the end.4installs28Event StudyWalks an R user through the full causal-inference event-study workflow on panel data: a three-step decision tree routes uniform treatment timing to a traditional TWFE event study with fixest and staggered adoption to five modern robust estimators, from Callaway and Sant'Anna's did package to Sun and Abraham via sunab(), Gardner's did2s, de Chaisemartin and D'Haultfoeuille, and Borusyak-Jaravel-Spiess imputation. Three reference files add joint pre-trend Wald tests, Bacon decomposition, placebo and permutation checks, and HonestDiD sensitivity analysis. Every plot follows the economics-journal standard the skill enforces: discrete point estimates with vertical 95% error bars, never connected lines with shaded ribbons.4installs29Bayesian WorkflowBayesian-workflow is an opinionated Bayesian modeling workflow built on PyMC and ArviZ. A developer consults it before writing Bayesian model code to apply guardrails that agents skip unprompted, like the nutpie sampler, prior and posterior predictive checks, LOO-PIT calibration, and non-centered parameterizations. It defines a ten-step sequence from formulating the generative story through reporting results. It emphasizes model criticism, convergence checks, and never reporting point estimates alone.3installs30Briefing NoteBriefing Note is a skill that generates structured 1-2 page policy briefing notes with issue, background, analysis, options, and recommendation sections. A government or consulting economist uses it to produce decision-focused documents in standard formats like UK GES or Australian Treasury. It can auto-populate with live data and import cost-benefit or input-output outputs.3installs31Causal InferenceCausal Inference is a skill for Bayesian causal effect estimation using PyMC, CausalPy, and DoWhy. A data scientist or economist uses it to answer 'does X cause Y' questions with methods like difference-in-differences, synthetic control, RDD, and instrumental variables. It enforces a DAG-first workflow, user checkpoints for assumptions, and mandatory refutation before any causal claim.3installs32Check Citationscheck-citations is a skill that verifies academic citations in .bib files against CrossRef, Semantic Scholar, and OpenAlex. A researcher uses it to catch AI-hallucinated, chimeric, and altered references before submitting a paper. It runs a Python script with JSON output and can be wired into pre-commit hooks or CI pipelines, requiring no API keys.3installs33Cost Benefitcost-benefit is a skill that runs cost-benefit analysis, producing economic and financial NPV alongside benefit-cost ratio, optimism bias adjustment, and wellbeing valuation. An economist uses it to appraise options against frameworks like the HM Treasury Green Book, EU Better Regulation, World Bank, and ADB. It adds sensitivity analysis, switching values, a validation gate, and a one-line headline verdict.3installs34DowhyDoWhy is a Claude skill for causal inference using Microsoft Research's DoWhy framework. A data analyst uses it to answer whether X causes Y by defining a causal graph, identifying the effect via backdoor/frontdoor/instrumental-variable criteria, estimating treatment effects, and validating with refutation tests. It fits observational-data analysis, A/B tests with confounders, and counterfactual reasoning.3installs35Econ AuditEcon Audit is a Claude skill that reviews economic analysis documents (fiscal briefings, macro briefings, market research) against methodology standards and practitioner best practice. An economist uses it to catch errors in counterfactual, additionality, discounting, double counting, and distributional analysis, and to detect strategic misrepresentation. It returns a red-amber-green scorecard with severity-ranked issues and an optional fix mode.3installs36Econ WriteThis skill is an economics paper writing assistant that synthesizes advice from 50+ guides by leading economists. A researcher uses it to write, edit, or restructure abstracts, introductions, results sections, and referee responses. It enforces concrete claims, active voice, and a reader-first structure.3installs37Fiscal BriefingThis skill produces a compact public-finances briefing for the UK, US, or Australia. An analyst uses it to summarize borrowing, debt, receipts, spending, and fiscal rules with every number traceable to an official source. It optionally adds a debt sustainability analysis via the debtkit R package and exports to markdown, xlsx, word, pptx, or pdf.3installs38LifelinesThis skill guides survival analysis in Python using the lifelines library. A data scientist uses it to fit Kaplan-Meier curves, run Cox proportional-hazards regression, and compare groups with log-rank tests on right-censored data. It is aimed at clinical-trial analysis, epidemiology, and prognosis modeling where time-to-event data must account for censoring.3installs39LonglistThis skill brainstorms a longlist of a project's benefits and costs for economic appraisal. An analyst uses it before a cost-benefit analysis to produce two clean tables, each item tagged with materiality, a cash-flow classification, and quantification and monetisation methods. It supports appraisal frameworks including the HMT Green Book, EU Better Regulation Guidelines, World Bank, ADB, and the Victorian Treasury HVHR framework, and hands off to a downstream business case.3installs40Macro BriefingThis skill produces an up-to-date macroeconomic briefing for the UK, US, Euro area, or Australia. It pulls live data on GDP, inflation, employment, wages, rates, trade, housing, and fiscal position from official sources and structures it following each central bank's reporting conventions. The output is a single one-page deliverable with a traffic-light macro assessment and traceable data sources.3installs41Market ResearchThis skill produces a structured, source-cited market research report for an industry or product. An analyst uses it to cover market sizing, key players, concentration (HHI/CR4), Porter's Five Forces, PESTLE macro-environment, regulation, supply chains, and outlook. It supports UK, US, EU, Australia, and global scope with multi-geography comparison, and cites every data point.3installs42OpenalexThis skill uses the openalex CLI to retrieve academic metadata from the OpenAlex API. A developer or researcher uses it to search papers, look up authors and institutions, track citations and references, filter and group works, and download open-access PDFs by DOI, ORCID, or OpenAlex ID. It matters for querying scholarly literature and metrics from the command line.3installs43Slr PrismaA skill that guides a user through writing a systematic literature review following the PRISMA 2020 reporting framework. It interviews the user, covers the full 27-item PRISMA 2020 checklist, produces a Word document manuscript in journal-article format, generates an annotated PRISMA flow diagram, and enforces APA 7th Edition referencing. A researcher uses it to structure or audit a systematic review. It explicitly does not cover meta-analysis or statistical pooling.3installs44Thematic AnalysisThis skill walks a user through conducting a thematic analysis of qualitative data using Braun and Clarke's 2006 six-phase framework. A researcher uses it to code interviews, focus groups, or open-ended survey responses into themes and write up a findings section. It outputs a Word document and an annotated thematic map, and does not cover IPA, grounded theory, or discourse analysis.3installs45StataA Stata reference skill for writing correct .do files covering data management, econometrics, causal inference, graphics, and Mata programming. Developers use it when asking an agent to write, debug, or explain Stata code. It documents Stata-specific pitfalls that cause silent bugs and includes routing tables plus command-line batch-mode execution instructions.2installs46StatspaiA skill for StatsPAI, an agent-native Python package for causal inference and applied econometrics exposing 390+ functions behind a single import and a unified API. Developers use it to run OLS, IV, DID, staggered DID, RDD, propensity score matching, synthetic control, double machine learning, causal forest, and neural causal models. It returns self-describing result objects with publication-ready Word, Excel, and LaTeX export.2installs47AbstractAbstract reads an academic manuscript and its registered notebooks to generate a structured abstract. A researcher runs it to draft or update the abstract of an empirical paper. It pulls the research question, data and methods, key quantitative results, and contribution, then targets a configurable word count. It writes the result into the abstract field of the manuscript's YAML front matter after review.1installs48Bayesian EstimationBayesian-estimation is a reference for Bayesian estimation and inference in quantitative social science. A researcher uses it when specifying priors, running MCMC, diagnosing chain convergence, or reporting posterior summaries, including hierarchical and structural models. It covers prior elicitation with weakly informative defaults, when to prefer Bayesian over frequentist methods, and debugging sampling pathologies. It spans tools like Stan, PyMC, NumPyro, brms, and rstanarm.1installs49Beamer PresentationBeamer Presentation Creator is a skill that builds academic slide decks in LaTeX Beamer. A developer or researcher uses it to turn a paper or project into conference, seminar, or job-market talk slides with a structured frame layout and clean theme. It provides duration-based slide structures and a ready-to-edit LaTeX template.1installs50C2C2 is a qualitative research design consultant skill for phenomenology, grounded theory, case study, narrative inquiry, and ethnography. A researcher uses it to choose a qualitative methodology after the research question is set, getting alternative design options scored by how conventional they are. It also produces participant selection strategy, sample size justification, and quality criteria.1installs51C3C3 is a mixed methods research design consultant skill that plans studies combining qualitative and quantitative approaches. A researcher uses it to pick among sequential explanatory, sequential exploratory, convergent parallel, embedded, and multiphase designs. It interprets Morse notation and specifies the integration strategy and workflow for each design.1installs52Causal Inference RCausal Inference in R is a skill that generates R econometric code for instrumental variables, difference-in-differences, and regression discontinuity designs. A researcher uses it to run treatment-effect estimation on panel data with proper diagnostics and clustered standard errors. It produces publication-ready code built on the fixest and modelsummary packages.1installs53Causal MlCausal ML is a skill covering causal machine learning methods for applied economics and social science. A researcher uses it to choose and implement estimators like double machine learning, causal forests, and meta-learners for high-dimensional controls or heterogeneous treatment effects. It covers cross-fitting and Neyman-orthogonal estimation using econml, DoubleML, and grf.1installs54Codebookcodebook is a skill that auto-generates a Markdown codebook documenting the variables in a dataset. A researcher uses it to produce a variable reference with types and summary statistics from a CSV, Stata, Excel, or Parquet file. It reads the data read-only and outputs a summary table plus per-variable detail sections with description placeholders.1installs55D2D2 is a Claude skill that designs and executes interview, focus group, and systematic observation protocols for social science research. A researcher uses it to build question guides, plan probing strategies, and set transcription conventions so qualitative data collection is rigorous and trustworthy. It distinguishes structured, semi-structured, and unstructured interview types and includes member-checking and confidentiality checkpoints.1installs56D4D4 is a Claude skill that develops measurement instruments (scales, questionnaires, surveys) for social science research. A researcher uses it to write items, choose response formats, run expert and cognitive review, and pilot test so the instrument has construct validity and reliability. It walks the full scale-construction process from conceptualization through psychometric analysis.1installs57Draft SectionDraft Manuscript Section is a Claude skill that expands bullet points or an outline into academic prose for a paper section. A researcher writing in Quarto uses it to draft two to five paragraphs in empirical economics journal style, wired with cross-references and citations from references.bib. It matches the manuscript's existing tone and inserts placeholders where a needed citation is missing.1installs58E2E2 is a Claude skill for systematic qualitative coding, codebook development, and theme identification. A researcher uses it to code interview or text data with deductive, inductive, or hybrid strategies, run thematic analysis, and assess saturation. It provides codebook templates, coding-by-methodology guidance, and CAQDAS pointers.1installs59E3E3 is a Claude skill for mixed methods integration, combining qualitative and quantitative data strands into meta-inferences. A researcher uses it to pick an integration strategy (connecting, merging, embedding), build joint-display matrices, and assess integration quality with legitimation techniques. It fits convergent, sequential, and embedded mixed methods designs.1installs60Econometrics REconometrics R is a Claude skill for R-based econometric analysis in academic research. A researcher uses it to run panel data, difference-in-differences, instrumental variables, spatial econometrics, and regression analysis with data.table, fixest, and sf, then produce publication-ready tables and figures. It includes modern DiD estimators like Sun-Abraham and Callaway-Sant'Anna and causal forests.1installs61Econ VisualizationThis skill generates publication-quality charts and graphs for economics papers. A researcher uses it to build line, bar, scatter, and event-study figures with consistent academic styling and export-ready PDF/PNG output. It defaults to R and ggplot2 with matplotlib/seaborn as a Python alternative.1installs62GeopandasThis skill covers spatial data analysis with geopandas and the broader Python geospatial stack. A researcher uses it for GeoDataFrames, spatial joins, CRS and projections, reading spatial file formats, choropleth and interactive maps, and spatial statistics via PySAL. It targets geopandas 1.x and organizes guidance through decision trees and reference files.1installs63Latex Econ ModelThis skill writes and typesets economic models in LaTeX with correct mathematical notation and academic conventions. An economist uses it for the theory section of a paper or for lecture notes, covering utility maximization, equilibrium conditions, dynamic programming, and game theory. It structures models into environment, preferences, constraints, equilibrium, and solution with example LaTeX output.1installs64Latex TablesThis skill generates publication-ready LaTeX tables for regression results and summary statistics using academic formatting. A researcher uses it to convert model output from Stata, R, or Python into clean tables with booktabs rules, significance stars, and notes. It standardizes table style across a paper and prepares tables for journal submission.1installs65LinearmodelsThis skill guides econometric estimation in Python using the linearmodels package. A researcher uses it to run panel-data models (fixed/random effects, between, first-difference, Fama-MacBeth), instrumental-variables and GMM estimators, and system regressions like SUR and 3SLS. It provides decision trees for choosing the right estimator and standard errors, and notes where pyfixest or statsmodels are better fits.1installs66Literature Gap FinderThis skill provides a framework for identifying research gaps in statistical methodology. A researcher uses it to build method-by-setting matrices, verify that a gap is genuinely unaddressed across databases, and score candidate gaps on impact, novelty, tractability, and other criteria. It is used when positioning a contribution or deciding on a research direction.1installs67Lit Review AssistantThis skill helps economists conduct literature reviews by structuring searches, summarizing papers, and synthesizing findings. A researcher uses it to define search terms across databases like EconLit, NBER, and Google Scholar, organize each paper into a structured summary, and identify agreements, disagreements, and gaps. It provides reusable templates for literature and per-paper summaries.1installs68Methods Paper WriterThis skill guides the writing of statistical methodology manuscripts for journals such as JASA, Biometrika, and Biostatistics. A researcher uses it to structure the paper, follow journal formatting requirements, apply a 6-paragraph introduction formula, and present simulation studies and theory with standard notation. It includes LaTeX templates and VanderWeele mediation notation conventions.1installs69Obsidian BasesThis skill creates and edits Obsidian Bases (.base files), which are YAML-defined database-like views over notes. A developer uses it to add filters, computed formula properties, and table, card, list, or map views, then validate the YAML and confirm it renders in Obsidian. It matters for turning a note vault into structured, queryable views.1installs70Obsidian CliThis skill uses the obsidian CLI to interact with a running Obsidian instance from the command line. A developer uses it to read, create, append, and search notes, set properties, list tasks and tags, and also to reload, debug, and screenshot Obsidian plugins and themes. It matters for scripting vault operations and plugin development workflows.1installs71Obsidian Experiment LogThis skill maintains canonical experiment and result notes in an Obsidian research vault. A developer or researcher uses it when a training run, ablation, evaluation, or finding changes the experimental state, updating existing notes rather than creating sprawl. It matters for keeping empirical-research records structured and linked from papers through to writing.1installs72Obsidian Link GraphThis skill repairs and strengthens Obsidian wikilinks among existing canonical project notes such as papers, knowledge, experiments, results, and writing. A developer uses it to improve navigation and backlinks and route new references to the best existing note. It matters for keeping a research vault connected without generating graph artifacts or note sprawl.1installs73Obsidian Literature WorkflowThis skill handles the literature sub-workflow inside a filesystem-first Obsidian research vault. A researcher uses it to ingest papers from Zotero into canonical Papers/ notes, normalize note structure, synthesize durable literature knowledge into Knowledge/ notes, and refresh a literature canvas. It matters for organizing paper notes and connecting them to project knowledge without an Obsidian MCP.1installs74Obsidian MarkdownThis skill creates and edits Obsidian Flavored Markdown, the Obsidian-specific extensions on top of CommonMark and GFM. A developer uses it for wikilinks, embeds, callouts, frontmatter properties, tags, comments, math, Mermaid diagrams, and footnotes when writing .md notes in an Obsidian vault. It matters for producing correctly-rendering Obsidian notes.1installs75Obsidian Project BootstrapThis skill bootstraps an Obsidian project knowledge base for the current repository. A researcher uses it to bind an existing code-plus-Markdown repo into a compact vault, running a detect step then a bootstrap script that creates hub, plan, and knowledge notes. It matters for starting a new research project's knowledge base without importing raw datasets or the whole code tree.1installs76Obsidian Project LifecycleThis skill changes the lifecycle state of an Obsidian project knowledge base using a shared helper script. A researcher uses it to detach, archive, or purge a project, and to archive, purge, or rename individual notes. It matters for cleanly retiring or reorganizing research projects while preserving or removing vault content as intended.1installs77Obsidian Project MemoryThis skill maintains a filesystem-first, agent-driven Obsidian knowledge base for a research project. A researcher uses it to detect and bind a repo, read minimum context, classify a knowledge delta, and write durable notes across papers, experiments, results, and writing without MCP or API keys. It matters as the main authority for keeping research project memory synchronized and durable.1installs78PlotlyThis skill is a reference for creating interactive data visualizations with Plotly in Python, using both Plotly Express and Graph Objects. A developer uses it for scatter, line, bar, histogram, box, heatmap, 3D, and geographic charts, plus subplots, styling, and HTML export, when hover, zoom, or pan interactivity is needed. It matters for producing web-based interactive charts in research and analysis pipelines.1installs79PlotnineThis skill is a reference for creating static, publication-quality figures with plotnine, a Python implementation of the grammar of graphics using ggplot2 syntax. A developer uses it for geoms, aesthetics, scales, coordinates, facets, and themes when static output is needed or coming from an R ggplot2 background. It matters for producing print-ready charts for reports and papers.1installs80PolarsA skill that guides an agent to write high-performance Polars DataFrame code in Python. It covers lazy and eager execution, the expression API, I/O for CSV, Parquet, JSON and databases, aggregations, joins, string and datetime operations, and pandas interop. A developer uses it when manipulating data with Polars, migrating a pipeline from pandas, or reading Parquet files. It matters because it steers the agent toward correct, performant patterns instead of row iteration.1installs81PyfixestA skill that guides an agent to run fixed-effects regressions in Python with pyfixest, the Python port of R's fixest. It covers OLS, Poisson, and IV estimation with multi-way fixed effects, difference-in-differences designs, clustered standard errors, wild bootstrap, and publication output such as etable tables and event-study plots. A researcher uses it when estimating panel models or DiD designs and needs publication-ready tables. It routes to linearmodels for random effects and statsmodels for GLM without FE.1installs82Referee ResponseA skill that drafts a structured, point-by-point response letter to peer-review referee comments. It parses each referee point, reads the manuscript to ground the reply, and writes a response plus a description of the changes made with section references. A researcher uses it after a revise-and-resubmit decision to prepare the response document. It follows academic conventions such as thanking the referee and quoting each comment.1installs83Regression TableA skill that formats estimation output from a notebook into a publication-quality regression table. It adds coefficient rows with standard errors, significance stars, fixed-effects Yes/No rows, and summary statistics like N and R-squared, following academic conventions. A researcher uses it when turning regression results into a results table for a paper. It recognizes statsmodels, linearmodels, R fixest, and Stata reghdfe output and emits a Quarto embed shortcode.1installs84ReplicateA skill that builds and verifies replication packages by routing to a reproducibility-auditor agent. The auditor runs structural checks on seeds, versions, paths, and pipeline integrity, plus functional checks on reproduction, data documentation, environment, and output matching. A researcher uses it before submitting a replication package to a journal to confirm results regenerate. It is a routing command that invokes the auditor agent on the current project.1installs85Reproducible PipelinesA skill that guides building reproducible research pipelines so every result is regenerable from raw data by running a single command. It covers standardized project directory structure, workflow managers (Make, Snakemake, DVC), computational environments (conda, Docker, renv), data versioning, and journal replication-package preparation. A researcher uses it when setting up a new empirical project or debugging works-on-my-machine failures. It matters because reproducibility is required by journal data editors.1installs86Robustness TableA skill that generates robustness-check code from a baseline regression and formats the results as a combined sensitivity-analysis table. It creates specifications such as alternative controls, alternative fixed effects, different clustering levels, subsample analysis, winsorizing, and placebo tests, then collects them into one publication-ready table. A researcher uses it to add sensitivity analysis to a paper. It recognizes statsmodels, linearmodels, R fixest, and Stata output.1installs87R Python TranslationA skill that maps R data-analysis packages and idioms to their Python equivalents for quantitative social science. It translates tidyverse, ggplot2, fixest, survey, sf, and plm to Python tools like polars, plotnine, pyfixest, statsmodels, linearmodels, and geopandas, covering data wrangling, regression, visualization, and causal inference. A developer with an R background uses it to audit or translate Python analysis code, or to annotate output with R-equivalent comments.1installs88Science CommunicationA skill that translates technical data science findings into clear communication for non-technical audiences. It covers audience analysis, narrative frameworks (Pyramid Principle, SCQA, AIDA), plain-language translation of statistical jargon, hedging and uncertainty, and deliverable formats like executive summaries and policy briefs. A researcher uses it when presenting findings to stakeholders or reviewing a draft for clarity. It focuses on what story to tell and to whom, not on building charts.1installs89Simulation ArchitectA skill for designing Monte Carlo simulation studies for statistical methodology research. It applies the ADEMP framework (Aims, Data-generating mechanisms, Estimands, Methods, Performance measures) from Morris et al. 2019, provides R code templates for parallel simulations, and guides on Monte Carlo standard errors and how many replications are needed. A methodologist uses it when designing a simulation study to evaluate estimators. It matters for rigorous, reproducible simulation reporting.1installs90Stata Data CleaningA skill that helps economists clean, transform, and prepare datasets for analysis in Stata with reproducible workflows. Developers use it when cleaning raw survey or administrative data, merging multiple sources, handling missing values or duplicates, and building analysis-ready panel datasets. It generates documented do-files with logging, assert checks, and codebook output for replication.1installs91Stata Python TranslationA translation reference that maps Stata commands, idioms, and workflows to their Python equivalents across data management, regression, causal inference, surveys, and visualization. Stata-background users use it to audit or learn Python analysis code, and agents use it to annotate Python output with Stata-equivalent comments. It also documents paradigm gaps where the two ecosystems diverge.1installs92StatsmodelsA skill for statistical modeling in Python with statsmodels, covering OLS/WLS/GLS, GLMs, logit/probit, count models, time series, mixed effects, diagnostics, and hypothesis testing. Developers use it when fitting regressions without fixed effects, running GLMs, or analyzing time series via the formula or array API. It routes to detailed reference files and flags where to use pyfixest or linearmodels instead.1installs93Structural ModelingA reference skill for building, estimating, and debugging structural econometric models. Researchers use it for BLP demand estimation, dynamic discrete choice (Rust, Hotz-Miller CCP), auction models, and any workflow with moment conditions, nested fixed-point algorithms, or MPEC formulations. It walks the full arc from economic model to moment conditions to estimated parameters and helps diagnose convergence failures.1installs94Submission PrepA skill that runs pre-submission checks on a journal manuscript and generates a submission checklist. Authors use it before submitting to verify word count, double-blind anonymization, citation resolution against references.bib, orphan placeholders, embedded figure/table sources, and cross-references. It reports each check with pass/fail status and leaves manual items unchecked for the user.1installs95SvyA skill for the svy Python package, which does design-based analysis of complex survey data. Researchers use it to specify survey designs (strata, PSU, weights, FPC), estimate means/totals/proportions with proper standard errors, run survey-weighted GLM regression, and do domain analysis for federal surveys like NHANES, CPS, and ACS PUMS. It is Polars-native and validated against R's survey package.1installs96Write WellThis skill checks the prose quality of Quarto (.qmd) manuscripts against William Zinsser's On Writing Well. It flags clutter, weak verbs, hollow qualifiers, cliches, inflated academic voice, and weak leads and endings. It writes a structured markdown report organized by document section and never edits the source file. It can be scoped to a single section via a @sec-label argument.1installs