expectations
Working expectations and documentation practices. Use when capturing learnings, documenting gotchas, recording architectural decisions, or understanding how to work with a codebase. Triggers on "document this", "remember this pattern", "what should I know about", or after completing significant features.
Works with
--- name: expectations description: Working expectations and documentation practices. Use when capturing learnings, documenting gotchas, recording architectural decisions, or understanding how to work with a codebase. Triggers on "document this", "remember this pattern", "what should I know about", or after completing significant features. license: MIT --- # Expectations ## When Working with Code 1. **ALWAYS FOLLOW TDD** - No production code without a failing test. Non-negotiable. 2. **Think deeply** before making any edits 3. **Understand the full context** of the code and requirements 4. **Ask clarifying questions** when requirements are ambiguous 5. **Think from first principles** - don't make assumptions 6. **Assess refactoring after every green** - but only refactor if it adds value 7. **Keep project docs current** - Update CLAUDE.md when introducing meaningful changes ## Documentation Framework **At the end of every significant change, ask: "What do I wish I'd known at the start?"** Document if ANY of these are true: - Would save future developers significant time - Prevents a class of bugs or errors - Reveals non-obvious behavior or constraints - Captures architectural rationale or trade-offs - Documents domain-specific knowledge - Identifies effective patterns or anti-patterns - Clarifies tool setup or configuration gotchas ## Types of Learnings to Capture - **Gotchas**: Unexpected behavior discovered (e.g., "API returns null instead of empty array") - **Patterns**: Approaches that worked particularly well - **Anti-patterns**: Approaches that seemed good but caused problems - **Decisions**: Architectural choices with rationale and trade-offs - **Edge cases**: Non-obvious scenarios that required special handling - **Tool knowledge**: Setup, configuration, or usage insights ## Documentation Format ```markdown #### Gotcha: [Descriptive Title] **Context**: When this occurs **Issue**: What goes wrong **Solution**: How to handle it // CORRECT - Solution const example = "correct approach"; // WRONG - What causes the problem const wrong = "incorrect approach"; ``` ## Code Change Principles - **Start with a failing test** - always. No exceptions. - After making tests pass, always assess refactoring opportunities - After refactoring, verify all tests and static analysis pass, then commit - Respect the existing patterns and conventions - Maintain test coverage for all behavior changes - Keep changes small and incremental - Ensure all TypeScript strict mode requirements are met - Provide rationale for significant design decisions **If you find yourself writing production code without a failing test, STOP immediately and write the test first.** ## Communication - Be explicit about trade-offs in different approaches - Explain the reasoning behind significant design decisions - Flag any deviations from guidelines with justification - Suggest improvements that align with these principles - When unsure, ask for clarification rather than assuming
More Writing & Documentation skills
paper-context-resolver
lllllllama/rigorpilot-skills
Rigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing README guidance by default.
repo-intake-and-plan
lllllllama/rigorpilot-skills
Rigor Intake helper for README-first deep learning repo reproduction. Use when the task is specifically to scan a repository, read the README and common project files, extract documented commands, classify inference, evaluation, and training candidates, and return the smallest trustworthy reproduction plan to the main orchestrator. Do not use for environment setup, asset download, command execution, final reporting, paper lookup, or end-to-end orchestration.
minimal-run-and-audit
lllllllama/rigorpilot-skills
Rigor Run skill for README-first deep learning repo reproduction. Use when the task is specifically to capture or normalize evidence from the selected smoke test or documented inference or evaluation command and write standardized `repro_outputs/` files, including patch notes when repository files changed. Do not use for training execution, initial repo intake, generic environment setup, paper lookup, target selection, hidden scientific-meaning changes, or end-to-end orchestration by itself.

