ai-codelisted

# Code ## Purpose Code implementation skill. Writes code that satisfies loaded context standards on the first pass. Lightweight self-review at build-time; full validation deferred to /ai-review. ## When to Use - New features and implementing approved plans - Adding functionality to existing modules - Writing utility/helper code NOT for: tests (use /ai-test), debugging (use /ai-debug), refactoring (dispatch `ai-simplify`), schema work (use /ai-schema). ## Process Step 0 (load contexts): read `.ai-engineering/manifest.yml` `providers.stacks`; load `.ai-engineering/overrides/<stack>/conventions.md` for each stack and `.ai-engineering/overrides/_shared/conventions.md`; load `.ai-engineering/team/*.md` for team conventions. ### Step 1: Pre-Coding Checklist Before writing any code: 1. **Restate the task** in one sentence -- confirm understanding 2. **Identify target files** -- existing files to modify or new files to create 3. **Search for existing patterns** -- grep for similar implementations in the codebase to match conventions 4. **Check decision-store.json** -- read `.ai-engineering/state/decision-store.json` for relevant architectural decisions ### Step 2: File Placement Protocol 1. New files go in the directory matching existing project structure -- follow the pattern, do not invent new paths 2. Test files mirror source structure (e.g., `src/foo/bar.py` -> `tests/foo/test_bar.py`) 3. Never create top-level files without explicit user instruction 4. If unsure abou