review-staged

A command that reviews staged Git changes before they are committed. Staged changes are edits already selected for inclusion in the next commit.

In plain words
What is it for?
Use it by running /review-staged to inspect staged files and their full diff through a multi-agent review process.
Why use it?
It helps find problems in the exact changes about to be committed and records review findings with file and line context. If nothing is staged, there is nothing for it to review.

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/microsoft/agent365-python/review-staged
Clone the repo
git clone --depth 1 https://github.com/microsoft/Agent365-python

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 1,573 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 $0.01573
Opus 5 $0.00000 $0.00787
Sonnet 5 $0.00000 $0.00315
Haiku 4.5 $0.00000 $0.00157

Measured yesterday against content hash 9217a0d2140f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review-staged 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 yesterday.

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/review-staged.md · 195 lines

How it starts

The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review Staged Changes

Review staged (uncommitted) changes using a comprehensive multi-agent code review process. Use this before committing to get feedback on your changes.

Usage

/review-staged

No arguments required - reviews all currently staged changes.

Instructions

You are coordinating a comprehensive code review for staged changes that haven't been committed yet.

Step 1: Gather Staged Changes Information

First, collect information about the staged changes:

  1. Get list of staged files:

    git diff --cached --name-only
    
  2. Get the full diff of staged changes:

    git diff --cached
    
  3. IMPORTANT: Save the diff output - you will need it to:

    • Include relevant diff context snippets in each finding
    • Determine the exact Diff Line values for inline comments (absolute line numbers in the target file; use the diff hunks to map changes to their file line numbers)
    • Determine the side (RIGHT for additions +, LEFT for deletions -)
  4. If there are no staged changes, inform the user and suggest they stage files with git add.

  5. Create the .codereviews/ directory if it doesn't exist.

Step 2: Launch Parallel Code Reviews

Launch THREE sub-agents in parallel using the Task tool. Each agent MUST receive:

  • The list of staged files (so they stay scoped)
  • The diff content for context

CRITICAL: You MUST launch all three agents in a SINGLE message with THREE parallel Task tool calls:

  1. architecture-reviewer (subagent_type: architecture-reviewer)

    • Prompt: "Review staged changes for architectural concerns. Files changed: [list files]. This is a pre-commit review, not a PR. Focus on design alignment, component boundaries, namespace patterns, and documentation gaps. Output your review in the structured markdown format specified in your instructions. Since this is not a PR, omit PR Link fields or mark them as N/A."
  2. code-reviewer (subagent_type: code-reviewer)

    • Prompt: "Review staged changes for code quality. Files changed: [list files]. This is a pre-commit review, not a PR. Focus on Python best practices, SDK usage, security, type hints, async patterns, and maintainability. Output your review in the structured markdown format specified in your instructions. Since this is not a PR, omit PR Link fields or mark them as N/A."

Read the full file on GitHub · 195 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. yesterday First seen · 195 lines · 0 tokens per session scan A 9217a0d2140f

Subscribe to this mod's changes

review-staged is a command published in the GitHub repository microsoft/Agent365-python (41 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,573 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.