code-review

A Git code-review command that asks several AI models to assess changes. Code review is the process of checking proposed code for security, quality, performance, and design problems.

In plain words
What is it for?
Use it to review recent commits, staged files, or a branch comparison, with attention focused on security, performance, quality, architecture, or all of these areas.
Why use it?
Using multiple reviewers can surface different risks in commits, staged changes, or the difference between two branches.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/osmantic/ods/code-review
Clone the repo
git clone --depth 1 https://github.com/Osmantic/ODS

Made for: Claude Code.

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 4,038 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00000 $0.04038
Opus 5 $0.00000 $0.02019
Sonnet 5 $0.00000 $0.00808
Haiku 4.5 $0.00000 $0.00404

Measured 2d ago against content hash 2d53b26c9a9e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-review scanned grade A with 0 findings 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 2d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/code-review.md · 494 lines

How it starts

The opening of the file, as written. The whole thing — 494 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Git Code Review with Multi-Model Consensus

Perform comprehensive code review using PAL MCP multi-model consensus. Reviews recent commits, staged changes, or branch comparisons with focus on security, quality, performance, and architecture.

Arguments

  • $ARGUMENTS - Review configuration:
    • commits=N - Number of recent commits to review (default: 5)
    • --staged - Review only staged changes instead of commits
    • --branch <target> - Compare current branch against target branch (e.g., main)
    • --focus <area> - Focus area: security, performance, quality, architecture, or full (default: full)
    • --post-pr - Post review as PR comment if in PR context

Philosophy: Ask Early, Ask Often

This skill should liberally use AskUserQuestion at every decision point. Code review is subjective — what matters depends on context, and assumptions about reviewer priorities are frequently wrong. The cost of asking is low; the cost of reviewing the wrong scope or missing what the user cares about is high. Specifically:

  • Before reviewing — confirm the review scope and what the user cares about
  • When the diff is large — let the user choose what to focus on first
  • When categorization is ambiguous — ask about severity levels
  • When models disagree — let the user break the tie on issue severity
  • After findings — ask what to do next (fix, re-review, post to PR)
  • When tradeoffs exist — present them explicitly instead of picking silently

The user should feel like they're directing the review, not receiving a fire-and-forget report.

Workflow

1. Parse Arguments

Extract configuration from $ARGUMENTS:

  • Default: commits=5, focus=full
  • Flags: --staged, --branch, --focus, --post-pr

Determine review mode (mutually exclusive):

Mode Trigger Description
Commits Default or commits=N Review last N commits
Staged --staged flag Review staged changes only
Branch --branch <target> Compare against target branch

Read the full file on GitHub · 494 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. 2d ago First seen · 494 lines · 0 tokens per session scan A 2d53b26c9a9e

Subscribe to this mod's changes

code-review is a command published in the GitHub repository Osmantic/ODS (5,145 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 4,038 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.