llm-review
01Plugin Claude Code
Plugin marketplace listing 1 plugin: llm-review.
Plugin Claude Code
Plugin marketplace listing 1 plugin: llm-review.
Plugin Claude Code
Vendor-neutral adversarial code review from Claude, Cursor, DeepSeek, Gemini, Kimi, and MiniMax — via API key or each tool's own CLI subscription. The codex-plugin-cc experience for every model.
Command
Guided setup for Claude reviews — choose API key or subscription (Claude Code CLI).
Command
Adversarial code review of your current changes by Claude, independent of this session (CLI subscription or Anthropic API).
Command
Guided setup for Cursor reviews — Cursor is CLI-only; auto-install cursor-agent and guide login.
Command
Adversarial code review of your current changes by Cursor (cursor-agent CLI, read-only).
Command
Guided setup for DeepSeek reviews — DeepSeek has no CLI, so this configures an API key.
Command
Adversarial code review of your current changes by DeepSeek.
Command
Guided setup for Gemini reviews — choose API key or subscription (Antigravity CLI agy, the successor to gemini-cli), auto-install the CLI if needed.
Command
Adversarial code review of your current changes by Google Gemini.
Command
List or inspect background llm-review jobs for this repo.
Command
Guided setup for Kimi reviews — choose API key or subscription (Kimi CLI), auto-install the CLI if needed.
Command
Adversarial code review of your current changes by Kimi (Moonshot).
Command
Guided setup for MiniMax reviews — MiniMax has no CLI, so this configures an API key.
Command
Adversarial code review of your current changes by MiniMax.
Command
Fan-out adversarial review — every working model (Claude, Cursor, DeepSeek, Gemini, Kimi, MiniMax) reviews your changes in parallel, then a consensus synthesis.
Command
Overview of which review providers/transports are ready, and how to configure each.
Command
Live connection status — actually probes each LLM provider to show what works in this local setup right now.
Skill Claude CodeCodex
Adversarial code review from multiple LLM providers (Claude, Cursor, DeepSeek, Gemini, Kimi, MiniMax). Use when the user asks for a second opinion, a code review from another model, an adversarial review, a multi-model or panel review, a sanity check on a diff, or mentions @llm-review or any provider name in a review…