--- name: "prog-expert-marimo" description: "Use when you need to create, edit, or debug Marimo reactive Python notebooks, including cell structure, reactivity rules, output display, and the Lint-Export-Assert test protocol. Expert-level programming and pattern management." metadata: stage: "alpha" source: "MIGRATED" requires: - "impl-experts-python-contract" --- # Prog Expert Marimo ## Goal Expert-level Marimo notebook development. Generates robust, reproducible `.py` notebooks with correct cell structure, reactivity rules, and self-verification using the Lint-Export-Assert protocol. ## Critical Knowledge (Always Apply) - **File format**: Pure `.py` only — never `.ipynb` - **Entry point**: `app = marimo.App()` + every cell is `@app.cell` - **Reactivity**: Variables returned from a cell are global; redefining a name in two cells is an error - **No side effects**: Never mutate; always produce new variables (e.g., `df_clean = df.dropna()` not `df.dropna(inplace=True)`) - **Output**: Last expression in cell auto-displays; use `mo.md(...)` for markdown; return figure objects (no `.show()`) - **Test protocol**: See `references/patterns.md` — Lint → Export → Assert before declaring done ## Core Workflow (Progressive Disclosure) 1. **Context Analysis**: Analyze the current codebase and project state. 2. **Knowledge Retrieval**: - Fetch latest library versions using `scripts/track_versions.py`. - Check `recipes/` for reoccuring patterns and implementation guides. - If needed, research new patterns via `scripts/research_knowledge.py`. 3. **Implementation**: Execute core logic. Leverage `atomic-examples/` and `recipes/` for guidance. 4. **Learning**: - Extract patterns from successful implementations via `programming/prog-expert-advisor/scripts/learn.py`. - Create or refine coding recipes in `recipes/` using `scripts/manage_recipes.py`. ## Instructions - ALWAYS use the preferred package manager for this ecosystem. - Before starting any task, search the `recipes/` directory for relevant implementation patterns. - If a task involves a recurring pattern not yet in `recipes/`, create a new recipe after successful implementation. - Proactively update your knowledge if library versions are outdated or new patterns are discovered. ## Self-Learning & Research - Gathers knowledge using web research and the `use-context7-api` skill. - Learn from successful implementations when used, from other code bases and from researching the documentation. - Keep track of the latest library versions. - Refine recipes to be more granular based on implementation experience. ## Auto-Improvement - Every time this skill is used, analyze the usage chat to find out if further improvement of the skill is advised. - Ask the user if those changes should be made. - If approved, store the improvement ideas in the `resources/improvement_ideas.md` file. ## References - [Patterns](references/patterns.md) — Reusable code patterns and best practices. - [Versions](references/versions.json) — Tracked library versions. - [Recipes](recipes/README.md) — Coding recipes for recurring tasks. ## Example Code When learning or implementing, use these code examples. ALWAYS load them via `view_file` to maintain Progressive Disclosure: - **Atomic Examples** (Small code chunks from docs): - *(Add examples here)*: `view_file(/atomic-examples/...)` - **Recipes** (Larger patterns): - **Notebook Template & Patterns**: `view_file(/recipes/recipe-marimo-notebook.md)` - **Reference Implementations** (Complex pseudo-code from existing real-world codebases): - *(Add references here)*: `view_file(/reference-implementations/...)` ## Constraints * Do not perform unauthorized or destructive actions. * Do not overwrite existing files without explicit user confirmation. ## Script Integration - **Research**: `uv run scripts/research_knowledge.py ""` - **Version Tracking**: `uv run scripts/track_versions.py` - **Recipe Management**: `uv run scripts/manage_recipes.py [args]` - **Example Management**: `uv run scripts/manage_examples.py [args]` - **Pattern Learning**: `uv run programming/prog-expert-advisor/scripts/learn.py `