← ClaudeAtlas

coderabbitlisted

Review pending changes with the CodeRabbit CLI: review the branch diff and working tree, fix findings judged real, dismiss false positives with justification, then run the quality gate. Use for CodeRabbit reviews of local changes.
NomadicDaddy/aidd · ★ 1 · Code & Development · score 60
Install: claude install-skill NomadicDaddy/aidd
# CodeRabbit Review Run the CodeRabbit CLI against the target application's pending changes, triage every finding, fix the ones that are real, and dismiss the rest with a written justification. CodeRabbit is an external cloud review service: the CLI sends the diff to CodeRabbit for analysis, so only use this skill on projects whose owner is comfortable with that. ## Usage ``` coderabbit [app] [base-ref] ``` If `[app]` is omitted, infer from the current working directory. `[base-ref]` is the branch to compare against; when omitted, use `main` if it exists (locally or as `origin/main`), otherwise `master`. ## Setup 1. Check if the CodeRabbit CLI is available: - Run `coderabbit --version` to test. - On Windows the official CLI does not run natively; if the native probe fails, try WSL: `wsl -e coderabbit --version`. If WSL has it, run every subsequent CodeRabbit command through WSL against the project's mount path, e.g. `wsl -e sh -c "cd /mnt/d/path/to/project && coderabbit review --plain"` (translate `D:\path\to\project` to `/mnt/d/path/to/project`). 2. If neither probe succeeds, STOP and report exactly what is missing instead of failing cryptically. Setup guidance to include in the report: - Install (macOS/Linux/WSL): `curl -fsSL https://cli.coderabbit.ai/install.sh | sh` - Log in: `coderabbit auth login` (interactive; the user must do this themselves) - On Windows, both steps happen inside WSL. 3. Verify the CLI is logged in