vibetunnel: Command for Claude Code

.claude/commands/review-pr.md

review-pr is a command for Claude Code from amantus-ai/vibetunnel. It costs 0 tokens per session (2,775 once invoked), scanned B, original, MIT.

A pull-request code review command that compares the current branch with the main branch and combines reviews from Claude and Gemini.

In plain words
What is it for?
Use it to review a pull request before merging, including focused checks for security or performance.
Why use it?
It helps find problems from more than one review perspective without requiring you to inspect every change manually. You can also narrow the review to security, performance, tests, architecture, or documentation.

Command for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents; mentions Gemini CLI.

This is amantus-ai/vibetunnel's own configuration. It tells Claude Code how to work on vibetunnel itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything vibetunnel configures →

About the project

VibeTunnel is a tool that displays a computer terminal in a web browser, allowing users to access terminal sessions and monitor coding agents remotely. It is for developers who want to check agents, builds, or shared terminal sessions from another device, and the catalogue contains instructions, commands, and an MCP integration for it.

amantus-ai/vibetunnel · 4,654 stars · on GitHub · vt.sh

Reuse

Borrowing it

Nothing to install: this file belongs to amantus-ai/vibetunnel. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/amantus-ai/vibetunnel/main/.claude/commands/review-pr.md
Clone the repo
git clone --depth 1 https://github.com/amantus-ai/vibetunnel

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for review-pr

README.md
[![agentmods](https://agentmods.dev/badge/commands/amantus-ai/vibetunnel/review-pr/github.svg)](https://agentmods.dev/commands/amantus-ai/vibetunnel/review-pr)
Your own site
<a href="https://agentmods.dev/commands/amantus-ai/vibetunnel/review-pr"><img src="https://agentmods.dev/badge/commands/amantus-ai/vibetunnel/review-pr/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for review-pr

Your own site · 80×15
<a href="https://agentmods.dev/commands/amantus-ai/vibetunnel/review-pr"><img src="https://agentmods.dev/badge/commands/amantus-ai/vibetunnel/review-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,775 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.02775
Opus 5 $0.00000 $0.01388
Sonnet 5 $0.00000 $0.00555
Haiku 4.5 $0.00000 $0.00278

Measured 12d ago against content hash 97bb734681f6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade B, and why

review-pr scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

touch .pr_review_diff.tmp && chmod 600 .pr_review_diff.tmp
.claude/commands/review-pr.md · 314 lines

How it starts

The opening of the file, as written. The whole thing — 314 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/review-pr

Perform a comprehensive code review of the current pull request using parallel sub-agents powered by Claude and Gemini, with results intelligently merged.

Usage

/review-pr [options]

Options:

  • --focus <area> - Focus on specific areas. Valid values: security, performance, tests, architecture, documentation, all
  • --gemini-only - Perform the review only with gemini CLI

Description

This command performs a thorough code review of the current branch's changes compared to the main branch. It runs two parallel sub-agents - one using Claude and another using Gemini - to perform independent comprehensive reviews. The results from both AI models are then intelligently merged to provide a complete multi-perspective analysis.

Examples

Standard PR review

/review-pr

Security-focused review

/review-pr --focus security

Performance and scalability review

/review-pr --focus performance

Implementation

When you use this command, I will:

  1. Initial PR Context Gathering

    • Check if required tools are available: gh, git, and gemini
    • Run gh pr view to check PR description and any existing comments (handle errors gracefully)
    • Run git log "main..HEAD" --oneline to see all commits in this PR (with proper shell escaping)
    • Run git diff "main...HEAD" --stat to get an overview of changed files (with proper shell escaping)
    • Generate comprehensive diff with secure temporary file handling:
      touch .pr_review_diff.tmp && chmod 600 .pr_review_diff.tmp
      git diff "main...HEAD" > .pr_review_diff.tmp
      
    • If any git commands fail, provide helpful error messages and guidance
  2. Launch Two Parallel Sub-Agents

    Sub-Agent 1: Claude Review

    • WARNING: This step must be skipped when --gemini-only has been passed
    • Performs complete multi-aspect code review using Claude
    • Analyzes all aspects below independently

    Sub-Agent 2: Gemini Review

    • Runs Gemini CLI following instructions from docs/gemini.md
    • Performs complete multi-aspect code review using Gemini
    • Form the prompt to the Gemini CLI in a way that it only returns the final output of its findings, to save tokens
    • Analyzes all aspects below independently
    • Use a timeout of 10 minutes for the gemini CLI command (configurable via environment variable GEMINI_TIMEOUT_SECONDS, default: 600)
    • IMPORTANT: Gemini CLI can only access files within the project directory
      • Create temporary diff file in project root (e.g., .pr_review_diff.tmp)
      • Use @.pr_review_diff.tmp syntax to include the diff in Gemini prompt
      • Clean up temporary file after review completes

Read the full file on GitHub · 314 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 314 lines · 0 tokens per session scan B 97bb734681f6

Subscribe to this mod's changes

review-pr is a command published in the GitHub repository amantus-ai/vibetunnel (4,654 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,775 tokens. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.