new-projectlisted

# Scaffold a new research project You are creating a new research project repository under `projects/`. ## Finding the workspace root The workspace root contains `CLAUDE.md` alongside `projects/`, `pipelines/`, `$ROOT/data_catalog/`, `research/`. If the current directory is inside a project or pipeline, search upward to find the root. Use `$ROOT` for all paths below. ## Step 1: Identify the project folder and read existing materials The user has already created a folder under `projects/<slug>/` and may have placed materials in it (PDFs, notes, descriptions, data files, drafts, etc.). 1. **Infer the slug** from the folder name. If the skill was invoked with an argument, use that as the slug. Otherwise, look for recently created folders under `projects/` that lack the canonical structure (no `docs/` directory). 2. **Read every file** in the project folder. These are the user's seed materials — project descriptions, research proposals, paper drafts, notes, data documentation, etc. 3. **Extract context** from these materials: - Research question and motivation - Current focus / what to work on first - Methods and identification strategy - Data sources - Relevant literature - Institutional background - Any other information that maps to the docs/ files 4. **Check workspace context** for connections: - `$ROOT/research/meta/data_linkages.md` and `$ROOT/data_catalog/` for existing datasets - `$ROOT/research/meta/variable_dictionary.md` for known varia