review

# Claude Code Review Use this skill when the user wants Claude Code to review the current working tree or a branch diff in this repository. Use `$cc:review` as the default when the user asks for code review, asks you to have Claude review something, or wants a second review pass without explicitly asking for stronger adversarial scrutiny. If the user asks for stronger challenge on design, tradeoffs, rollout risk, migration risk, configuration behavior, or provides custom review focus text, route to `$cc:adversarial-review` instead. If the user wants Claude Code to investigate, validate by changing code, or actually fix/implement something, route to `$cc:rescue` instead. If the overall request is "you review it too, also ask Claude to review in the background, then you aggregate and fix it", keep the delegated Claude part on `$cc:review` unless the user explicitly asks for a harsher or more adversarial review. `$cc:review` does not accept custom focus text. If the user wants to steer Claude toward a particular angle, question, subsystem, or risk area, that is a signal to use `$cc:adversarial-review` instead. Resolve `<plugin-root>` as two directories above this `SKILL.md` file. Always run the companion from that active plugin root: `node "<plugin-root>/scripts/claude-companion.mjs" review ...` Supported arguments: `--wait`, `--background`, `--base <ref>`, `--scope auto|working-tree|branch`, `--model <model>`, `--effort <low|medium|high|xhigh|max>` (defaults: model=opus and...