Run a local PR review without the @claude GitHub Action.
Scanned 9/5/2026
Install to Claude Code
npx -y skills add bravros/bravros --skill local-review --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Local Review?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/bravros-local-review-a6113208)More formats (shields.io, HTML) on the badges page.
---
name: local-review
description: Run a local PR review without the @claude GitHub Action.
---
Read [briefing.md](references/briefing.md) on demand for detailed context and instructions.
# local-review
INTENT: the same review `@claude` does, fully local — a FRESH zero-context subagent reviews the diff; findings are saved, posted to the PR for audit-trail parity, and routed.
HARD CONSTRAINTS:
- **The orchestrating session MUST NOT write the review itself** — dispatch ONE `Agent` (`code-reviewer`).
- **This skill never writes `.planning/.review-stamp-*.json`.** `bravros pr-review --write-stamp` is the ONE stamp-write authority.
Flags: `<PR>` · `--deep` · `--post`/`--no-post`.
## Overview
1. **Gather**: `gh pr view "$PR"` + `gh pr diff "$PR"` + file stats.
2. **Dispatch**: Send diff and repo checks to subagent (`sonnet` / `opus` on `--deep`).
3. **Parse + save**: Extract verdict block and save `.planning/pr-reviews/${PR}-<TS>.md`.
4. **Post + route**: Comment on PR (unless `--no-post`) and route to `/finish` or `/address-pr`.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.