reviewcodelisted

# Code Review Review the diff for correctness bugs and cleanups: $ARGUMENTS This is a port of Claude Code's built-in `/code-review`, which is user-invocable only and cannot be reached through the Skill tool. This one is model-invocable, so `ship`, `review:peer`, and any other skill can delegate to it. The upstream text it was ported from is in [references/upstream-2.1.215.md](references/upstream-2.1.215.md). ## Arguments - **Effort level**: the first token, if it matches `^(low|med|hig|xhi|max)[a-z]*$` case-insensitively. Prefixes count: `med`, `hi`, `xh` all resolve. An unrecognized level-shaped token is ignored with a note (`Ignoring unrecognized effort "<token>"; valid: low, medium, high, xhigh, max. Using <level>.`) and is not treated as a target. - **`--fix`**: apply findings to the working tree after reporting. May appear anywhere. - **`--base <ref>`**: review against this base instead of the resolved default. - **`<target>`**: everything else, free-form. A PR number, branch, ref range, path, or a plain-English scope restriction ("only `src/parser.ts`", "focus on error handling", "skip the test churn"). - **`ultra`**: not supported here. Stop and tell the user to type `/code-review ultra` themselves. It launches a billed cloud review and only a user-typed invocation can start it. With no effort level, use the session's effort. Default to `medium`. ## Phase 0 — Scope Resolve the diff: 1. If `--base <ref>` was passed, the range is `<ref>...HEAD`. 2. Otherwise `git