What problem does it solve? Turning an empirical social-science paper into reusable, structured research-design knowledge is slow and error-prone; this Skill converts one paper into linked wiki pages covering variables, datasets, models, mechanisms, identification, robustness, and heterogeneity. ## Core Features & Use Cases - Structured Paper Ingestion: Reads a local PDF or TeX source, prepares it with tools/prepare_paper_source.py, and extracts only text-supported empirical facts such as sample periods, measurement formulas, fixed effects, and clustered standard errors. - Wiki Page Generation: Creates or updates pages under papers/, variables/, datasets/, models/, mechanisms/, hypotheses/, identification/, robustness/, heterogeneity/, and tables/ using the runtime page templates. - Knowledge Graph Edges: Adds typed edges (operationalizes, uses_dataset, estimates_model, tests_mechanism, addresses_endogeneity_with) via tools/research_wiki.py, then rebuilds index.md, context_brief.md, and open_questions.md and lints the wiki. - Use Case: A researcher downloads a Chinese-language PDF of an empirical economics paper with no online metadata and ingests it into the project wiki, producing a paper card plus linked variable, dataset, and identification pages without inventing any unreported details. ## Quick Start Run /empirical-ingest with the path to your local paper PDF or TeX file, optionally adding --topic to tag the research topic.
